Valuing infrastructure projects

From time to time an infrastructure project or its [potential future] shareholders may need to know the value of the project’s equity. Generally this will be for the purposes of a sale or purchase, or possibly for a listing.

It is most common to use a discounted cash flow methodology when conducting such a valuation. Most of the time this is fairly easy, as the project is likely to have a project model which produces distributions to shareholders at the bottom of the cash flow waterfall.

However, there are a few considerations to take into account when conducting a valuation:

Starting balances

Unless you are conducting the valuation as at financial close, the project will have built up certain actual balances during the construction period and any operations period that has elapsed. These will include the likes of cash in bank, debtors, creditors and plant and equipment. It is best to source these from the most recent set of financial accounts or management accounts available, as these will be the most accurate figures available. These balances will need to be incorporated into the project model such that in due course they feed through to the cash flow waterfall accurately. For example, a debtor balance will, via the model’s working capital assumptions, feed through into the cash flow waterfall and ultimately to distributions.

Starting balances, even cash balances, should not be included as stand-alone sums in the valuation. This is because the timing of distributions creates a material difference to the valuation, and generally any balance lacks this information unless it feeds through the waterfall. For example, a cash balance may result from a debt covenant lock up, which may only be cured in several months’ time.

More information about incorporating balances into the project model is available here: Operationalising Project Models.

Input assumptions

There are three main sources of assumptions available to the valuer:

  • The Project Documents and the Finance Documents. Good examples of such assumptions are base tariffs prior to inflation and interest margins. These assumptions will, all things being equal, hold for the life of the project, and consequently the valuer should leave them unchanged in the base case project model.

 

  • Expert information. Certain technical and macroeconomic assumptions may have been set by experts at the beginning of the project life or subsequently. An example of such would be future fx or inflation assumptions. These are best sourced from a (reputable and justifiable) expert source. Some technical assumptions may not change considerably over the life of the project, and may therefore be left alone without significantly effecting the valuation. An example of this might be power plant capacity.

Note that some project models may have a lenders’ base case and an equity base case. Typically it is most appropriate to use the equity base case (eg P50 energy yield assumptions) since it is the distributions which are being valued.

 

  • Actual experience. The performance of the project may differ to the original expected performance, in which case, in the absence of a reason to the contrary, the assumptions should be updated to reflect the most recent experience. Experience should be of reasonable duration – 12 months might be a good starting point to incorporate a reasonable sample size and any seasonality effects. Non-contracted opex inputs are a good example here, as is recent technical performance. Obviously when updating assumptions to reflect recent performance, one needs to apply a reasonability filter. Updating opex or technical assumptions to reflect an abnormally good or bad year will create bias in the valuation. One also needs to ensure that the experience is adjusted to account for the model’s inflation from the base date of the assumptions to the present.

A critical valuation assumption which deserves particular mention is the risk discount rate used to discount the distributions back to the valuation date. There are different means of arriving at, and influences upon, the appropriate discount rate for the valuation, and often some judgement will be required. These may include the following:

  • If the valuation is part of a sale negotiation, the risk discount rate may have been negotiated in advance, with the monetary valuation flowing from the risk discount rate as a result;
  • There may be recent sales which provide discount rate market benchmarks;
  • The project may have been structured so as to arrive at a reasonable risk discount rate for the risk profile of the project, as at the date of development;
  • The project may have experienced certain milestones and events which increase or decrease risk. An obvious one of these is the project completing construction and reaching Commercial Operation Date successfully. Another is the acquisition of an operating track record of a statistically significant sample length;
  • For completeness, there are theoretical models for determining an appropriate risk discount rate using the market risk free rate and the project’s beta and leverage – the Capital Asset Pricing Model (CAPM). Such a model is perhaps less relevant for an infrastructure project than for a normal operating company.

If a particular assumption has considerable uncertainty and has a significant effect on the distributions of the project – eg if plant output is very uncertain in the future for some reason – then it may make sense to model that assumption into the project’s future explicitly, rather than by modifying the risk discount rate to reflect the additional uncertainty. This allows greater understanding of the change in the valuation resulting from poor performance relating to the assumption in question. A good example of an assumption like this may be the traffic experience in a toll road project.

 

Valuation date

The date of the valuation may have considerable impact upon the valuation. In particular, moving the valuation date later by a month will have the following effect:

  • The valuation will increase by approximately 1/12th of the discount rate; and
  • The valuation will decrease by any distribution that was in the valuation period before the date change, but is no longer in the date change.

Consequently, if distributions are semi-annual, the valuation can move as much as 41% of the discount rate from a 5-month change in valuation date.

 

Termination amount

If one simply discounts the distributions in the project model, it is possible to miss the residual value of the project at the end of the project term (usually the end of the offtake contract term). This may be completely appropriate, for example in the case of a Build Own Operate Transfer project. However, in other instances it may not be appropriate at all, such as in the case where it may be feasible to enter into another subsequent offtake contract, or dispose of some of the project’s assets for a sum. Conversely, a rehabilitation cost may be applicable at the end of the term. Such an amount may be modeled explicitly by discounting the sales price or rehabilitation cost, or by extending the term of the modeled operations period using a new set of (possibly conservative) assumptions.

 

Sensitivities

The valuer may be asked to provide a single valuation figure, but it is nonetheless good practice to provide sensitivities to those assumptions which either:

  • Have considerable uncertainty;
  • To which the valuation outcome is very sensitive; or
  • Both.

The purpose of the sensitivities is to help the client in any final negotiations or decisions around the final price or valuation. If the final report is a public document, or shared with the other side of the negotiating table, these sensitivities could be removed from the final public version of the valuation report.