Quantifying the Unquantifiable: TWDB’s Proposed SWIFT Rule and the Challenges of Governing through Point Systems

This summer, the Texas Water Development Board (TWDB) proposed a rule to implement the state’s new $2 billion water infrastructure financing scheme.  The statutory framework governing the scheme is elaborate but, as is often the case with legislative enactments, somewhat general in parts, leaving many key decisions to the TWDB.

The proposed rule fills in some of – but not all – of the contours.  Notably, it does not address with specificity the way that financings will be structured.  (In the preamble to the proposed rule, the TWDB writes that “very few decisions on the structure and terms of financing, beyond what is set out in HB 4, are made in these proposed rules … The board has tentatively decided to not place those suggestions in the proposed rule. The board’s current opinion is that the terms and structure of SWIFT and SWIRFT will of necessity need to change over time. In order to preserve the ability of the board to respond as quickly as events dictate, such as changing market conditions and varying demands for funding, the board is opting to keep as much flexibility as possible with the board by keeping the rules on structure and terms of the SWIFT and SWIRFT to a minimum.”)

Of what is included in the proposed rule, the most intriguing relates to the process for prioritizing infrastructure projects seeking financial assistance.

Statutory Point System Requirement

The enabling legislation includes a provision, now codified at Texas Water Code § 15.437, that directs the TWDB to “establish a point system for prioritizing projects … The system must include a standard for the board to apply in determining whether a project qualifies for financial assistance at the time the application for financial assistance is filed with the board.”

Section 15.437 further requires that, in establishing this point system, the TWDB must give the “highest consideration” in awarding points to projects that will have a “substantial effect,” including projects that will: “(1) serve a large population; (2) provide assistance to a diverse urban and rural population;(3) provide regionalization; or (4) meet a high percentage of the water supply needs of the water users to be served by the project.”

In addition, Section 15.437 mandates that the TWDB must “also consider” seven other factors that presumably do not indicate that a project will have a “substantial effect” but that nonetheless advanced preferred public policies.  These second-tier factors include, among others, whether there is an emergency need for the project and whether the project would contribute to water conservation.

The enabling legislation does not interpret the first-tier or second-tier factors or discuss the relative weights or interactions among them.  It leaves all such architectural questions to the TWDB to address through rulemaking.

Points under Proposed Rule

The proposed rule would award predetermined amounts of points based on first-tier and second-tier factors (i.e., a project would receive five points if its applicant would be funding at least half of the project costs).  It would assign points on the basis of twelve factors – that is, the first- and second-tier factors set forth in the statute.

In designing the point system,  however, the board staff had to reckon with additional questions that the statute left unanswered, such as: where to set the break points within each factor; the relative points within factor; and the relative points across factors.

To take one example, Section 363.1304(5) of the proposed rule awards points to projects that would “meet a high percentage of the water supply needs of the water users to be served.”  To that end, it would distribute points as follows:

  • at least 50 percent of needs met, 10 points; or
  • at least 75 percent of needs met, 20 points; or
  • at least 100 percent of needs met, 30 points; or
  • less than 50 percent of needs met, zero points.

Meanwhile, another factor – found at 363.1304(12) – would give points based on how regional planning groups (RPGs) ranked the projects in their regional prioritization process:

  • top 80 percent of regional project ranking, 3 points; or
  • top 60 percent of regional project ranking, 6 points; or
  • top 40 percent of regional project ranking, 9 points; or
  • top 20 percent of regional project ranking, 12 points; or
  • top 10 percent of regional project ranking, 15 points; or
  • less than 80 percent of regional project ranking, zero points.

Under this design, serving at least 100 percent of needs would be worth three times as much as serving only 50 percent of needs.  It would also be worth twice as much as being ranked within the top 10 percent of an RPG’s prioritization.

Across all prioritization factors, the proposed rule awards the most points for the factors/break points shown in the table below.  (For a comparison of all the factors/break points under the proposed rule, please see this spreadsheet.)

If applicants receive the same scores, the conservation factor will serve as the tiebreaker.  If applicants receive the same scores on the conservation factor, the emergency need factor will serve as the backup tiebreaker.

50 Point Limit

For the first-tier factors, applicants can receive up to 50 points; for second-tier factors, there is no upper limit on points.

Reaching the 50 point threshold should not be a particularly uncommon feat.  All projects that “either directly, or in conjunction with other recommended water management strategies” will serve more than 1 million people receive 30 points.

Because utilities provide water services through interconnected water systems, a particular amount of water that a particular segment of the service population receives will seldom be directly traceable to a particular infrastructure project.  Instead, a project will increase overall system capacity and all customers will benefit equally from the marginal increase in system capacity.  In this way, even a small project could serve a population of more than 1 million so long as it connects to a system that serves a population of that size.

As a result, all projects submitted for SWIFT funding from applicants who serve populations of one million or more would automatically earn 30 points, more than half of all the points that they would be eligible to earn.

In a state with 6 of the 20 most populous cities in the country, and 2 of the 10 most populous metro areas, the million-person threshold is readily attainable.  Any project put forward by the cities of Dallas or Houston would notch 30 points just for showing up.

Point Systems as Policy Structures

The comment period on the proposed rule runs until September 1, and stakeholders will probably weigh in on the appropriateness of the break points and the particular point values assigned.  A discussion of such issues is beyond the scope of this blog post, though it is worth noting that TWDB undertook extensive outreach and data-gathering efforts as its staff drafted the proposed rule.

At a more abstract level, using point systems to distribute limited public resources is far from unusual.  Governmental entities have used them for such varied purposes as awarding affordable housing tax credits, college admissions, and port inspections.  The point systems help to guard against arbitrariness decision-making; yet they can also produce a kind of arbitrariness by favoring projects that happen to perform well in an automated exercise.

The point systems that best straddle this line are calibrated so that they incorporate and successfully promote policy priorities rather than devolving into collections of bureaucratic checkboxes.  But structuring a point system requires assigning numerical values to policy preferences that are somewhat soft and amorphous.  It requires quantifying that which is not perfectly quantifiable.

Why should a project that would serve 20,000 people and another that would receive 200,000 people receive the same number of points, as they would under the proposed rule?  Why should a project that would serve 1 million people receive 5 times more points than both?  Why should a project that would serve 2 million people receive the same number of points as a project that would serve 1 million?

These are policy questions that would perhaps be implicit in considerations of individual applications but that the point system makes explicit.  To some extent, the development of the point system forces the TWDB – and its stakeholder community – to address these questions now, systematically, at the rulemaking phase rather leaving them to the TWDB staff and board members to ruminate over in a more ad hoc way in the course of allocating financial assistance.

Point Systems and Discretion

HB 4 restructured the water board and put it under the control of three full-time board members appointed by the governor.  The bill requires that “[o]ne member must have experience in the field of engineering, one member must have experience in the field of public or private finance, and one member must have experience in the field of law or business.”

The care that the bill invests into matters of governance reflects on the significance of the board members’ perspectives and judgments.  And by virtue of their offices, the board members are certainly qualified to consider the merits of competing projects.

The point system, however, attempts to reduce decision-making to a fixed formula, at least in part.  The influence that the point system will have remains uncertain since the proposed rule only requires that TWDB’s executive director use it to prioritize applications; it does not address if the board must abide by this prioritization when awarding financial assistance or can simply approach it as a suggestion.

Energy Center Summer Intern Christine Wozniak (USC Gould School of Law ’15) assisted in researching this post.


Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>