Using real options in evaluating PPP/PFI projects

Transcription

1 Using real options in evaluating PPP/PFI projects N. Vandoros 1 and J.-P. Pantouvakis 2 1 Researcher, M.Sc., 2 Assistant Professor, Ph.D. Department of Construction Engineering & Management, Faculty of Civil Engineering, National Technical University of Athens, Bld. of Strength of Materials - Annex A, 9 Iroon Politechniou st, Athens, Greece. jpp@central.ntua.gr Abstract: The appraisal of PPP/PFI projects relies on financial evaluation methods such as the Net Present Value analysis. However, this method sometimes fails to model uncertainties, irreversibility of decisions and managerial flexibility and may, therefore, lead to wrong decisions. Newer methods, such as the real options analysis may assist in this direction by taking into account that a project may be delayed, abandoned or sub-divided strategically into components. Some or all of the above changes may, in turn, reduce uncertainty or otherwise benefit the project realization process. These arguments are further elaborated in this paper by the use of a hypothetical toll road project which is studied comparatively by both traditional and options methods. It is concluded that real options, despite the difficulty in their proper application, may provide better decision-making for evaluating PPP/PFI projects at the appraisal stage. Keywords: PPP, PFI, Real Options, Evaluation. 1. Introduction Public Private Partnerships (PPP) or Private Finance Initiative (PFI) projects have gained popularity worldwide because of the shortage of public sector financial resources and the acknowledgement of certain private sector advantages. At the same time, these projects are difficult to design, implement and operate due to the complexities, interdependencies and uncertainties involved and the different perceptions of the related issues between their stakeholders. A large amount of initial investment is required whereas the payback period is long and uncertain. For this reason, PPP projects must first prove their financial viability so that their implementation is endorsed. Traditional evaluation methods, such as the Net Present Value analysis fail to evaluate them properly, since their special characteristics, namely risks, government guarantees, negotiations and project financing cannot be taken into account (Ho and Liu, 2002). Various evaluation methods have been developed to enhance the traditional methods. One of them is real options. Real options analysis facilitates flexibility in management decisions, especially in later stages of implementation which are due to unexpected events. The use of real options gained popularity in the construction industry. Some of the researchers who used the real option approach are mentioned hereafter. Sing (2002) appraised the value of time-to-build option in a commercial development project. Yeo and Qiu (2003), 594

2 acknowledged the need for managerial flexibility and used real options analysis. Nordvik and Liso (2004) applied the real option methodology to show the interdependencies between the climate change and the behaviour of building owners. Ekstrom and Bjornsson (2005) used the real option approach to evaluate architecture, engineering, and construction information technology investments. Yiu and Tam (2006) applied the real option concept in the bidding process. However, research has also been conducted in the use of real options for the evaluation of both PPPs and conventional projects. Rose (1998) examined and evaluated two options for a toll road, the concession period option and the deferral of the concession fee and acknowledged that the options can help in the proper estimation of the value of project. However, it was also observed that the interactions of the options have an impact on the value of the project. Ho and Liu (2002) developed a real option pricing model taking into account the uncertainties concerning the project net cash flow and the construction cost. They arrived to the conclusion that their model constitutes a basis for PPP project financial evaluation. Ford et al. (2002) used a binomial option pricing model to represent alterations in design and arrived at the conclusion that the option can enhance managerial flexibility. Garvin and Cheah (2004) illustrated the potential of the option to defer evaluation of an infrastructure investment. According to them, options are easy to apply but their assumptions must be seriously taken into account. Huang and Chou (2006) combined the option to abandon and the minimum revenue guarantee in the pre-construction phase of a BOT infrastructure project, and arrived at the conclusion that real options analysis may offer distinct benefits. In any case, options combination may have a negative impact on their values. Mattar and Cheah (2006) introduced a new category of risks, which they called private risks and investigated the effect of this new risk category on real options. They ended up that their category of risk could be considered as the premium of the real option analysis. The intention of this paper is to examine how the real options analysis can improve the financial evaluation of a PPP project. For this reason, a comparison has been made between the traditional evaluation methods and the real options analysis. A difference between the current work and the previous ones is that the comparison is not made purely between Net Present Value (NPV) and real options but sensitivity analysis and scenario analysis are also performed. Another difference is that two out of three real options presented in this paper are not based on mathematical formulae, but on a way of thinking. It is revealed that real option analysis is not always mathematical, it is also, and this is more important, a conceptual process. 595

3 Table 1: Previous work on real options for PPP project evaluation Research Rose (1998) Evaluation of two options: concession period option deferral concession fee option Ho and Liu (2002) Development of an option pricing model, incorporated two variables: project net cash flow and construction cost Ford et al. (2002) Evaluation of a binomial option pricing model to design flexibility Evaluation of an option to defer, an economically corrected decision-tree Conclusions Options can help in estimating the project value, but their interactions have an impact on it The option model provides an adequate framework for the PPP financial evaluation Real options can boost the project flexibility Garvin & Cheah (2004) Option models provide simplicity and applicability but their assumptions must be taken into account Huang & Chou (2006) Combination of two options: Option approach looks option to abandon minimum promising but the revenue guarantee options combination decreases their values. Mattar and Cheah (2006) Real option and private risks Private risks can be considered as the premium of the real option analysis. 2. Traditional evaluation methods The traditional evaluation methods rely on the discounted cash flow series for the calculation of the Net Present Value (NPV). Their considerable advantages are the clear, consistent decision criteria for all projects, which are relatively simple, widely taught and accepted. Also, they are not influenced by the risk preferences of the investors. Finally, their results are easy to understand and interpret. A positive NPV means that the project is financially viable. On the other hand, these methods do not take into account the dynamic environment of a project and the continuously changing reality of business; therefore they are outdated (Yeo & Qiu 2003). Using these methods, the investor has no other choice than take it or leave it. However, it is very hard to make a decision when there are uncertainties and risks involved. This is the case with the PPP projects. As time progresses, new information is available through market research or feasibility studies. These changes may require flexible project design. The traditional evaluation methods do not properly depict these changes. Additionally, the NPV method is sensitive to the discount rates. A small increase or decrease in the discount rate will have a considerable effect on the final output of NPV. Many of the above mentioned disadvantages of the NPV are mitigated through the sensitivity analysis as well as the scenario analysis. 596

4 3. Real option analysis Contrary to the common belief, real options concept is not new. The first reference to real options is found in the Aristotle s writings. He described how Thales the Melesian, an ancient philosopher, concluded from tea leaves that there would be an olive harvest in six months time. Then, he agreed with the owners of olive presses to buy the right to rent their presses at the usual rate. When the harvest was collected, he rented the presses at above the market rate to the growers who were in need, while he paid the normal rate to their owners. From the above reference, it is clearly illustrated the concept of real options. Options are known from the financial world where they represent the right (not the obligation) to buy or sell a financial asset at a predetermined price (the exercise price) at the end of a predetermined period. The selling or buying of an asset at the predetermined price is termed exercising the option. For a European type of option, it is only possible to exercise the option at the end of the option period. On the contrary, it is possible to exercise the option at any time using the American type of option. Real options are based on the same principles as financial options. Having a real option means to have the possibility for a certain period to choose either for or against something, without binding up front. This possibility provides flexibility in the decision process. But real options differ from the financial ones, in terms that the real options are concerned with the strategic decisions of an organization. Consequently, they need to be considered from a wider viewpoint, while financial options can always be used independently. The most common real options are the option to defer, to abandon, to alter, to switch and the growth option. Real options gained popularity through the work of Black and Scholes (1973). Their formulas for the prices of a European call option (c) and a European put option (p) on a non-dividedpaying stock at time zero are: rt c = S N d ) Xe N( ) (Equation 1) 0 ( 1 d 2 p = Xe rt N( d 2 ) S0 N( d1) (Equation 2) where: d 2 d 1 ln( S = 0 2 / X ) + ( r + σ / 2) T σ T 2 ln( S0 / X ) + ( r σ / 2) T = = d1 σ σ T T while N(x) is the cumulative probability function for a variable that is normally distributed with a mean of zero and a standard deviation of 1.0, S 0 is the stock (current) price at time zero 0, X is the strike price, r is the continuously compounded risk-free rate, σ is the stock price volatility and T is the time during which the option is exercised. 597

5 4. The hypothetical toll road project A hypothetical toll road infrastructure project is used to compare and evaluate the traditional evaluation methods against the real option analysis. A public authority, wanting to develop a region s infrastructure, has decided to construct a toll road under the PPP scheme. The concession period is decided to be 30 years, while 5 years is estimated to be the construction period and the cost of the works. It is assumed that the initial traffic volume will be vehicles per day, and it is expected to grow at a rate of 10% in years 7-11 and 5% in years The initial toll is anticipated to be 2,00 and it will grow in accordance with the inflation. Specifically, due to the expecting good course of the economy, the inflation estimation is 3,1% for years 7-10, 3,0% for years 11-14, 2,9% for years 15-18, 2,8% for years 19-22, 2,7% for years and 2,6% for years The operating expenses are estimated to be 65% of the income for the first year of operation (year 6). The operating expenses are estimated to grow due to the maintenance works, 5% for the years 7-11, 7% for the years 12-16, 9% for the years 17-21, 10% for the years 22-26, 12% for the years The discount factor is estimated to be 15%. The estimations and assumptions mentioned above are summarised in Table 2. Table 2: Estimations and assumptions of the hypothetical toll road for NPV method NPV Initial estimation Fluctuation Variations Period Concession period 30 years Construction period 5 years Investment cost Discount factor 15 % Traffic volume vehicle/day Increase 10% 7-11 years 5% years Toll 2 Decrease 3,1% 7-10 years (year 6) 3,0% years 2,9% years 2,8% years 2,7% years Operating expenses 65% of the income (year 6) 2,6% years Increase 5% 7-11 years 7% years 9% years years years 4.1 NPV analysis using Discounted Cash Flow. Using this analysis, the resulted NPV is , thus it is proved that the project is not financially feasible. The NPV calculations of the hypothetical toll road are tabulated in Table

6 Table 3: Estimations and assumptions of the hypothetical toll road for sensitivity analysis and scenario analysis Sensitivity analysis Variations Traffic volume ± 20 % Operating expenses ± 20 % Discount factor ± 20 % Investment factor ± 20 % Scenario analysis Probability Variations Normal 50 % Optimistic 30 % Traffic volume + 30 % Pessimistic 20% Traffic volume 30 % 4.2 Sensitivity analysis After the calculation of NPV, a sensitivity analysis is performed. Four variables, namely traffic volume, operating expenses, discount factor, and the investment cost are selected to fluctuate 20%, so the sensitivity of calculated NPV to these changes is noted (Table 3). Table 5 presents the possible NPV for each variable alteration. It is observed that the critical variable is the discount factor. It is confirmed the above-mentioned disadvantage of the NPV. Moreover, all the calculated NPV is negative with only two exceptions, which means that the project is still not financially viable. 4.3 Scenario analysis The next step was the creation of scenarios. In a toll road, the critical variable is the traffic volume that is the generated income. So, three possible scenarios were created and a subjective estimation of the occurrence of each scenario was made. Specifically, an optimistic scenario is estimated to be that the initial traffic volume will be 30% higher than the calculated and the probability of this occurrence is assigned to be 30%. A pessimistic scenario is considered to be that the actual traffic volume will be 30% less than the calculated one with 20% probability of occurrence (Table 3). NPV for each of the scenarios are calculated and an Expected NPV is calculated to be (Table 6). Therefore, the project still does not seem to be viable. 599

7 Average Traffic Volume Toll Table 4: NPV calculations of the hypothetical toll road Gross Revenues Operating Expenses Operating Income Investment Cost Discount Factor Discounted Income Discounted Investment Cost 1 0 0, , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , Total NPV Table 5: Sensitivity analysis NPV +20% Normal -20% Traffic Volume Operating expenses Discount factor Investment cost

8 Table 6: Scenario analysis NPV Normal 50% Optimistic 30% Pessimistic 20% Total NPV Real option analysis Although the NPV, the sensitivity and scenario analyses show a significant negative amount, it is expected that the investment decision can be better improved through the real option analysis. Hence, a real option analysis is performed. Three real options to abandon are examined because these types of options can create substantial values in a project (Huang and Chou, 2006) Option to abandon (1) In many cases, the investor is provided with the possibility to abandon the project in a specific time. In this study, the investor can abandon the project after 1 year. That is, the investor has the right (not the obligation) to abandon the project without cost after 1 year. This procedure can be simulated with a European call option. Using the above-mentioned Black-Scholes formula (equation 1), the price of the option is calculated. The calculations and the assumptions are presented in Table 7. S 0 is the present value of gross revenues and X is the present value of total costs, where total costs are the sum of the operating expenses and the investment cost. The variable r is estimated to be 3% while the volatility is anticipated to be 25%. Finally, the time to exercise the option is 1 year. The price of the calculated option is ,83. Comparing to the traditional NPV analysis, the actual value of the process is [ ,83 + ( )] = ,83. Comparing to the scenario analysis, the actual value of the process is [ ,83 + ( )] = ,83. Equivalent calculations can be made between the sensitivity analysis and the real option. Table 7: Real options assumptions and analysis Present Value Of Turnover (S 0 ) Present Value Of Total Cost (X) Riskless rate 0,03 Volatility (σ) 0,25 Time (T) 1 Option Price ,83 601

9 d1-0,42 d2-0,67 N(d1) 0,34 N(d2) 0, Option to abandon (Passive process) As mentioned before, the real options analysis is not based only on a set of equations but it also involves a decision making process. For instance, the investor may decide to wait for one year in order that the uncertainties become resolved and the market revealed. If the optimistic scenario occurs, the investor will go ahead, without any cost; otherwise he may abandon the project. Hence the payoff of the options is given by payoff optimistic scenario NPV EURO = normal and pessimistic scenario abandon Option to abandon (Active process) Instead of waiting passively for the uncertainties to become resolved and the market revealed, the investor can decide to adopt a more rigorous process and conduct a market research. Market research is supposed to take 6 months to be conducted. In this case, if the market research signifies a favorable condition, that is the optimistic scenario will take place, then the payoff of the options is given by optimistic scenario NPV = EURO payoff = cost of market research + 6 normal and pessimistic scenario abandon months of operation The downside of this process is that there is a market research cost (it can be said that it represents the premium paid to obtain the option), while the profit is that the project will start 6 months earlier than in the previous case. 5. Conclusion The financial viability of a PPP project is vital for its implementation. The traditional evaluation methods, such as NPV, cannot properly assess the dynamic environment of these kinds of projects. This paper presents a comparison between three real options and the NPV traditional evaluation method by the use of a hypothetical project. Sensitivity analysis and scenario analysis were also performed to overcome the demerits of NPV. The comparisons show that the real options method can indeed enhance the value of a project. Nevertheless, real options cannot be seen as a panacea. Some of the drawbacks found in the NPV process are also incorporated, such as the accuracy of the cash flow. The actual 602

10 advantage of the real options analysis is that it can depict in a more clear way the decision making process which takes place in a PPP evaluation procedure. That means, that this kind of analysis cannot be based exclusively on mathematical equations; it requires analytical thinking as well. The use of a hypothetical project instead of a case study could be considered as a limitation of the current work. However, the primary purposes of this study were to indicate how the real option approach enhances the value of a project and to show the related analytical mental process. These purposes have been fulfilled. The use of a case study in future research could be a more stable and rigid basis for conclusion drawing. Further research could also incorporate the game theory so as the interaction between participants can also be taken into account. Thinking in terms of real options and game theory is believed to boost the decision making process which takes place in a PPP evaluation procedure. References Black, F., Scholes, M., (1973), The pricing of options and corporate liabilities, Journal of Political Economy, Vol. 81, No. 3, pp Ekstrom, M. A., Bjornsson, H. C., (2005), Valuing flexibility in Architecture, Engineering, and Construction Information Technology Investments, ASCE Journal of Construction Engineering and Management, Vol. 131, No. 4, pp Ford, D. N., Lander, D. M., Voyer, J. J., (2002), A real options approach to valuing strategic flexibility in uncertain construction projects, Construction Management & Economics, Vol. 20, No. 4, pp Garvin, M. J., Cheah, C. Y. J., (2004), Valuation techniques for infrastructure investment decisions, Construction Management & Economics, Vol. 22, No. 4, pp Ho, S. P., Liu, L. Y., (2002), An option pricing-based model for evaluating the financial viability of privatized infrastructure projects, Construction Management & Economics, Vol. 20, No. 2, pp Huang, Y.-L., Chou, S.-P., (2006), Valuation of the minimum revenue guarantee and the option to abandon in BOT infrastructure projects, Construction Management & Economics, Vol. 24, No. 4, pp Mattar, M. H., Cheah, C. Y. J., (2006), Valuing large engineering projects under uncertainty: private risk effects and real options, Construction Management & Economics, Vol. 24, No. 8, pp Nordvik, V., Liso, K. R., (2004), A primer on the building economics of climate change, Construction Management & Economics, Vol. 22, No. 7, pp Rose, S., (1998), Valuation of interacting real options in a toll road infrastructure project, The Quarterly Review of Economics and Finance, Vol. 38, Special Issue, pp Sing, T. F., (2002), Time to build options in construction processes, Construction Management & Economics, Vol. 20, No. 2, pp Yeo, K. T., Qiu, F., (2003), The value of management flexibility - a real option approach to investment evaluation, International Journal of Project Management, Vol. 21, No. 4, pp Yiu, C. Y., Tam, C. S., (2006), Rational under-pricing in bidding strategy: a real option model, Construction Management & Economics, Vol. 24, No. 5, pp