Public-Private Partnerships and the Proposed Indianapolis Justice Center

Interesting article in yesterday’s Indy Star about the $1.75B justice complex being proposed for Marion County: Weighing the scales: $1.75B justice center could be a bargain or a boondoggle.

The article in the Indy Star is less about the justice center itself (which nearly everyone seems to agree is needed) than about the procurement method being selected to build and operate it — so-called performance-based infrastructure. This is a form of public-private partnership (often called P3) that will be familiar to Monroe County residents, because it is being used for construction and operation of I-69 Section 5 (I have written about P3 used for Section 5 here). It has already been used as well for the Ohio River Bridges project near Louisville (and note that this method is NOT what was used with the Indiana Toll Road).

Per the studies cited in this article, the results of performance-based infrastructure (PBI) have been mixed. However, the most interesting aspect to me from a public policy perspective is how dependent these cost-benefit analyses (comparing the performance-based infrastructure procurement method against the traditional method, where the government bids out the construction, bonds for the funding, and then operates the facilities) are on the assessment of risk. Depending on how much risk you build into your cost-benefit model, the savings from performance-based infrastructure can be made to appear much larger or much smaller. This is because one of the primary benefits from PBI is that the contractor absorbs cost and schedule overruns (along with other risks), not the government. However, how much actual risk is avoided can often (always?) be a matter of disagreement and dispute.

This also leads to another question: how do you fairly assess the success of such a project after the fact? Once the project is completed, the risk that was avoided is obviously no longer a factor. But did the government still get the benefit of avoiding a risk of an event that wound up not happening anyway? Did you get the benefit of having car insurance last year even though you didn’t have any accidents or making any claims?

In any case, the article is worth reading. We will undoubtedly be seeing more, not less, of these P3/PBI deals, as local and state governments continue to try to build and operate infrastructure with flat or diminishing tax revenues.

Time Value of Money Analysis in Calculating the Costs and Benefits of Large Projects

Screenshot 2014-03-13 19.14.02Time Value of Money

This posting is going to be a little bit different from my typical blog postings, in that I’m going to write about a more general analytic concept that is particularly important to keep in mind when trying to understand projects that are either paid for or receive revenues over a period of time — a concept that is often called the “time value of money“.

This posting is prompted by many comments that I have heard from friends, colleagues, fellow public officials, and other community members in reaction to the controversial decision by the Indiana Finance Authority to award a 35-year contract to design, build, operate, maintain, and finance section 5 of I-69, a 21-mile segment of I-69 from south of Bloomington to south of Martinsville — a so-called Public-Private Partnership, or P3.

Please note that the following discussion does not in any way address the issue of whether building the highway at all is a good investment by the state. All it does is address the issue of paying for a large project like this over time vs. paying for it with current revenues/cash on hand.

I-69 Example

Without getting too far into the weeds on the specifics, the arrangement is that the State of Indiana will pay a contractor $21.8M per year for 35 years for the design, construction, operation, maintenance, and financing of the 21-mile section of highway. The government estimate of the cost of the design and construction of the highway was around $350M. However, if take the $21.8M per year times 35 years, you get a total of $763M. The claim that has been made, therefore,  is that by using this P3 financing vehicle in which payments are made to the contractor over time, that the government is paying more than double for the highway than it would if paid through conventional means (i.e., up front, with cash).

However, this conclusion ignores the time value of money (TVOM).  The basic principle behind TVOM is that a given amount of money today is worth more than it is at some future point. When you think about it, this should be self-evident; just ask yourself: given the choice, would you prefer $100 to be given to you right now or in a year? Once you accept that premise, the next question is how much more is that given amount of money worth today than it is at a future point? The answer is: a given amount of money is worth more today than it is at some future point by the amount you could earn by investing (or otherwise using) that given amount of money until that future point.

In order to compare alternative investments (or financing schemes) that are made over various periods of time, we need to make sure we compare apples to apples. One way to do that is to convert all alternatives to present value (PV) and then compare. Let’s consider our I-69 example. In that example, the state will be paying the contractor $21.8M per year over 35 years. However, the $21.8M in year 2 is not worth as much as the $21.8M in year 1, and the $21.8M in year 35 is certainly not worth as much as it is in year 1. So how do we convert the entire 35-year payout to Present Value, as though it were all being paid out immediately? And more importantly — how do we compare the 35-year payout against a $350M cost if the design and construction were paid out of cash today.

Basically, we discount future payments by the amount of money we could earn on the money we save by not having to pay it this year! How much do we discount it by — in other words, how do we compute the present value?

Quick Mathematical Interlude

I’m going to take a moment to do a little math here; however, if you want to skip to the next section, you won’t lose much of the overall argument. The actual equation is: PV = FV / (1+i)^n, in which PV is Present Value, FV is Future Value, i is the interest rate per period that you could earn on the money, and n is the number of periods that you could earn that interest rate. For a quick example, let’s consider the following alternatives:

  • Alternative A: I give you $100 today
  • Alternative B: I give you $100 a year from today

First of all, we know that alternative A — the $100 I give you today has a present value of $100.To to compare it to alternative B (I give you $100 a year from now), though, we need to compute the present value today of the $100 I pay you a year from now. In the above equation, $100 is the future value (FV) — the amount that the $100 paid in one year will be worth at the time it is paid. i is the interest rate that we could be earning per year (or per any time period) with the $100.  This is where TVOM analysis gets a little squishy, and the results you get can differ a lot depending on your assumptions. How much money can you earn with $100 in a year? Depends a lot on how you invest it! If in a savings account, almost nothing — less than 1%. However many investments earn quite a bit more than a savings account. One number that is considered pretty fair to use is the average municipal bond rate. At the very least, the municipal bond rate can be considered a good proxy for the opportunity costs of the money over a given period of time.

The following Web site provides municipal bond rates for various maturity ranges and credit ratings:

Just for the sake of this example, I’m going to take a national rate for a 10-year bond with AAA credit rating (which Indiana has) — an interest rate of 2.20% per year (of course, this amount may change).

Going back to our equation, we have: FV = $100, i = 0.022 (the 2.2% interest rate), and n=1 (1 year). The present value of that $100 paid out in a year is: PV = 100/(1+0.022)^1 = 100/1.022 = $97.85. In other words, with the assumptions we made, $100 paid to you a year from now is only worth $97.85 today.

I-69 Example, Revisited

OK, so let’s apply the TVOM principle to analyzing the 35-year contract for Section 5 of I-69.  For the purposes of illustrating TVOM in comparing the 35-year contract with a conventional financing (i.e. paying for the project out of current tax receipts and cash balance), I am simplifying the situation dramatically in the following way: the 35-year $21.8M/year payout to the contractor does not only include the design and construction of the road, but also the operations and maintenance of the road — money that would have had to be spent anyway in all 35 of the outyears regardless of the method of financing the design and construction of the road. If we really want to compare alternatives fairly, we would subtract out the costs of maintaining and operating the road for all 35 years — figures I don’t have close to hand. So this TVOM analysis can really be considered to be a worst-case from the perspective of the P3 scenario. In other words, the real cost of the P3 versus conventional financing is much more in the favor of P3 than it is in the following numbers.

I created the following table for all 35 years of payments (all numbers are in millions). The first column, Payment, shows the actual payment made to the contractor each year for the 35 year period of performance. The following 5 columns show how much those 35 years of annual payments are worth today (the only fair way to compare the arrangement against a conventional financing arrangement where the whole thing is paid with cash/current taxes), using 5 different interest rate assumptions (1%, 2%, 3%, 4%, and 5%). For a project of this size, given the municipal bond rates for 30 year bonds for AAA credit, the most realistic assumption is probably somewhere around 4% (from the recent past history of municipal bonds, probably a little under 4%).

Time Value of Money I-69 Example
Time Value of Money I-69 Example

So with the 4% interest rate assumption, we can see that the present value of 35 years of $21.8M annual payments is not $763M (i.e. 35 times $21.8M), but the much lower $406.89M. Obviously that number changes depending on the interest rate assumption. The higher the interest rate, the lower the present value. This should make intuitive sense: the more opportunity I have to make use of the money up front, the less valuable it is for me to have the money in the future compared to the present.

So to get back to our hypothetical-not-so-hypothetical example. We want to compare paying for a $350M highway construction project out of current dollars against the 35-year $21.8M/year P3 arrangement. With our assumption of 4% interest rate, the present value (cost) of the 35-year arrangement is $406.89M, compared to $350M if paid in cash — more, to be sure — but dramatically less than simply adding up the 35 annual payments ($763M).  And again, this doesn’t even include the fact that some of the annual payments cover the costs of maintenance and operations of the highway, which would be paid out anyway, regardless of how the design and construction are financed.

Conclusion

 None of this discussion should be taken as an endorsement of the P3 process, or the fact that the P3 financing model basically guarantees that the winning contractor will be a large multinational corporation with deep access to finance. But I do want to make sure that as we move forward with an arrangement that by all accounts will only become more common, discussions of these kinds of arrangements are based on facts. And the fact is that a given sum of money is less valuable in the future than it is today.  Most certainly we will be paying more for the privilege of stretching out the payments for the road over 35 years. But we won’t be paying double…not even close.

Winning I-69 Proposal Available from Indiana Finance Authority

On Friday, I wrote about the winning and losing proposals for I-69 Section 5 (the section from Bloomington to Martinsville, IN):

Today, the Indiana Finance Authority released the winning proposal (well, most of it — some parts are redacted) on its Web site. The proposal is in a number of separate files. To find it, go to the Indiana Finance Authority I-69 Web site and scroll down to “Selected Proposer”.

Right off the bat, based on a very quick read, I would think that the parts of the proposal most interesting to the public include:

  • Executive Summary
  • Financial Proposal Volume I
    • The section labeled “Type and Purpose of Each Funding Source and Facility” details the sources of funding used for the project as well as the overall costs of building and financing. I’ve included a screen shot below.
    • The equity member (prime contractor), Isolux Infrastructure, is committing $44.75M of its own investment to build the highway. They also plan to raise $253.51M through private activity bonds (PABs). Commitment of PAB funding by the underwriters is provided in this volume as well.
    • The “Uses” column on the right outlines how the funds would be spent. The first three items (Construction Costs, Construction Oversight Costs, and Operations & Maintenance Costs during Construction) add up to the $325M in construction costs that has been quoted in press releases.
  • Technical Proposal Volume 1
    • This section details the team’s approach to construction, design, operations and maintenance of the highway. A lot of this volume is very high-level and not particularly specific. Most of the pages consist of required letters of authority, certifications and representations, references (the contents of which have been redacted!), responsible proposer forms, etc. However, there are a couple of factors of note in the volume:
    • Page 7 includes a high-level Gantt chart (which I screen-shot below) outlining the design and construction schedule, indicating a construction start in mid-late 2014 and a finish by the end of 2016 (Volume 2 of the technical proposal, page 20, refers to a deadline of October 31, 2016 as the”Baseline Substantial Completion” deadline).
    • Rehabilitation of the pavement is scheduled for years 15 and 30.
  • Technical Proposal Volume 2
    • Volume 2 contains a wealth of information about the proposed approach to all aspects of design and construction, including pavement, bridge structures, bicycle and pedestrian access, drainage, lighting, traffic signals, etc., as well as communications and public outreach — far too much to summarize here.  There are just a couple of elements I quickly wanted to call out.
    • Page 43 provides a more detailed schedule of key completion milestones:
    • Page 47 indicates that the proposal design specifies that the construction will be asphalt, rather than concrete. Both were acceptable in the Request for Proposals; sections 1-4 are being constructed using concrete. Per this proposal, bridge structures, retaining walls, and noise walls will be constructed using concrete. As mentioned above, rehabilitation of the asphalt pavement is scheduled for years 15 and 30. Assuming that the asphalt pavement has a 15-year useful life, the rehabilitation at year 30 will mean that the pavement will have 10 years left of useful life when contract is completed (year 35).
    • Screenshot 2014-02-23 12.07.07

Undoubtedly there is much more to comment on in the thousands of pages of this proposal; however, I wanted to get a couple of quick highlights out to the public as soon as possible.

Winning and Losing Teams for I-69 Section 5 Released

logoYou have probably already heard in the media about the selection of a winning proposal to design, build, operate, maintain, and finance Section 5 of I-69. However, you probably haven’t seen all of the winning and losing teams yet!

This past Wednesday, the Indiana Finance Authority preliminarily announced the winning bid to design, build, operate, maintain, and finance the construction of Section 5 of I-69, 21 miles of highway from Bloomington to Martinsville. I-69 Development Partners, led by prime contractor Isolux Infrastructure Netherlands B.V. from Spain, was selected as the preferred proposal.

The winning proposal would design and build the highway for $325M. The state will pay $21.8M per year over a period of 35 years (minus any penalties for non-performance), which will cover not only the design and construction, but also all maintenance and operation, including snow and ice removal, repair, resurfacing etc. for the entire 35-year term of the contract. In the form of public-private partnership used for this contract, the contractor provides the financing for the project. However, unlike most arrangements in which the contractor provides the financing for an infrastructure project, this project will not be a toll-road.

The full press release from the Indiana Finance Authority can be found here:

More interesting than the press-release, however, is the full list of proposing teams, including all of their subcontractors.  In all, there were 4 teams proposing, all with very generic names: Connect Indiana Development Partners, Plenary Roads Indiana, WM 1-69 Partners, LLC, and I-69 Development Partners. Each team consists of an “equity member” (essentially a prime contractor) and over a dozen partners and subcontractors, including construction, design, environmental, operations and maintenance, etc.

The actual contract has not yet been released to the public; however, according to the Indiana Finance Authority, “portions of the preferred proposal” will be posted on its website next week. You can be sure that MoCoGov will be watching!