Key Messages 
  • Investment decisions on adaptation actions are surrounded by significant uncertainty. Approaches considering flexibility and future learning are becoming increasingly recognised as valuable tools to support decision-making.
  • Real Options Analysis (ROA) aligns with the concepts of iterative adaptive (risk) management, providing a means to undertake economic appraisal of future option values through the value of information and learning, and the value of flexibility, under conditions of uncertainty.
  • Even if ROA is a relatively complex tool to use, which requires a significant amount of information, it can be useful as a guide to some of the key parameters in public-sector decision-making for adaptation.

Context

Adaptation must be dynamic because preferences may vary with time and new or improved climate information and technologies may become available. Taking into account flexibility (e.g. of infrastructure design) and future learning (e.g. improved knowledge) into adaptation strategies can be very valuable to support decision-making under uncertainty (Chambwera et al., 2014).

Real Options Analysis (ROA) can help consider and value flexibility and learning into the economic assessment of adaptation. ROA has evolved from the financial economics literature and is intended to deal with future uncertainties of a project’s implementation (Zeng and Zhang, 2011). In the context of adaptation economics, it can be said that “ROA quantifies the investment risk with uncertain future outcomes” (Watkiss et al., 2015: 407). “This includes the flexibility over the timing of the capital investment, but also the flexibility to adjust the investment as it progresses over time, i.e. allowing a project to adapt, expand or scale-back in response to unfolding events. The approach can therefore assess whether it is better to invest now or to wait – or whether it is better to invest in options that offer greater flexibility in the future” (Watkiss et al., 2015: 407).

This Insight presents the steps undertaken to apply ROA to inform decision-making on a public investment in infrastructure planned to reduce flood-risk in the city of Bilbao (Basque Country, Spain), which involves opening a pre-existing canal that will turn the current peninsula of Zorrotzaurre into an island in the Bilbao Estuary. The interest of this infrastructure lies in the fact that a new urban development has been approved for the district of Zorrotzaurre, which is a flood-prone area in the Bilbao Estuary. As a response to the strong concern of the Basque Water Agency (URA) in relation to the development of a new urban district in an area subject to severe risk of flooding, the option of opening of the Deusto canal, which would turn Zorrotzaurre into an island significantly reducing the risk of flooding upstream, was put forward for consideration.

Policy and methodological developments 

A stochastic approach to damage estimation

The first step in the ROA is to estimate expected damages at different points in time. To do this a stochastic function was developed, using damage data from a previous study (Osés Eraso et al., 2012) as an input. The stochastic function considered two variables: the frequency of extreme flood events and the stochastic growth rate of damage, as a function of climate change and socio-economic development. This function enabled the calculation of expected flooding costs for any given time, depending on the difference between the discount rate and the sum of the increase of damages due to climate change and economic growth.

Using the stochastic damage function, the expected flood damages of different return periods and expected benefits (in terms of avoided impacts) of the opening of the Deusto canal were estimated (Figure 1).

Figure 1. Net present value of the damage for the baseline (Zorrotuaurre as a peninsula) and the adaptation scenario, once the opening of the canal is finished and the Zorrotuaurre district is an island.

Including uncertainty and risk into damage values

The second step is to estimate two risk measures that have proven to be very useful in contexts of uncertainty, namely the Value-at-Risk (VaR) and the Expected Shortfall (ES). The first one is the most standard measurement and well recognised by international financial regulatory bodies. In the case study, the VaR of damage resulting from river flooding in the Bilbao Estuary expresses the losses that could occur with a given confidence level α of 95%, for a time interval of 85 years.

The second risk measure is the Expected Shortfall (ES), which in this case represents the expected damage when VaR is exceeded, that is, the expected damages that would occur in the worst 5% of the cases. ES is, therefore, a better measure of risk for low probability but high damage events and a more robust indicator to assess risk (Rockafellar and Uryasev, 2002). Both measures of risk have been estimated for the Bilbao case study, as the opening of the canal is expected to reduce not only the average expected damage but also the level of risk. The risk assessment was performed using Monte Carlo simulations for the baseline, where Zorrotzaurre is a peninsula, and for the adaptation scenario once the Deusto canal is opened.

The MonteCarlo method is needed to estimate the distribution of damages probabilities for different events so that we can calculate the probability of exceeding different levels of damages.  However, it is important to check that the method yields results consistent with those generated by the underlying stochastic process for flood occurences. 

Figure 2 shows the reduction of risk, measured as ES(95%), for the baseline (peninsula) and canal opening (island) cases.

Figure 2. Representation of the Expected Shortfall (ES(95%)) for the baseline and opening scenario as a function of ρ-μ.

The damage distributions generated so far increase deterministically over time. However, the damage resulting from a given flood is not fixed but could vary for a number of reasons. One step further in the treatment of uncertainty would be to incorporate stochastic damage distributions. In the case of Bilbao, the stochastic model developed enabled us to consider that risks grow as volatility increases, though with an expected value identical to the case of deterministic growth. In other words, risk is not only a function of the discount rate and the economic growth, but also depends on volatility.

Options to invest and applying ROA to the data

This step aims to evaluate the economic impact of different investment timing. In calculating the timing of investments through ROA, there are various parameters which influence the maximum cost that can be accepted for making investment immediately. For example, volatility can change the boundary of the wait-investment regions. As shown in the next figure, the greater the volatility, the lower the investment boundary. In other words, higher volatility makes potential investors more demanding and they invest only when the cost is lower.

Figure 3. Investment and wait regions depending on volatility.

Main implications and recommendations 

The application of ROA demonstrates that it is a relatively complex tool to use, which requires a significant amount of information. The risk estimation can be carried out using tools that are now available, but in the climate adaptation context is not always the case. The example of flood prevention is a public good, which holds for most public adaptation projects. Decisions of timing can be informed by using decision trees, in which the concepts of value at risk and expected loss play a central role. But including many decision points generates very complex decision trees, which are difficult to communicate to policy-makers. ROA is useful as a guide to some of the key parameters in public-sector decision-making for adaptation but may not be suitable for use as a standard tool that can be applied in a straightforward manner.

Bibliography 

Chambwera, M., Heal, G., Dubeux, C., Hallegatte, S., Leclerc, L., Markandya, A., McCarl, B.A., Mechler, R., Neumann, J.E., (2014), Economics of adaptation, in: Field, C.B., Barros, V.R., Dokken, D.J., Mach, K.J., Mastrandrea, M.D., Bilir, T.E., Chatterjee, M., Ebi, K.L., Estrada, Y.O., Genova, R.C., Girma, B., Kissel, E.S., Levy, A.N., MacCracken, S., Mastrandrea, P.R., White, L.L. (Eds.), Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp. 945–977.

Osés Eraso, N., Foudi, S., Galarraga, I. (2012), Análisis del impacto socio económico del daño por inundación en la Ría de Nervión: un cambio de escenario ante la apertura del Canal de Deusto. Informe de Avance del Proyecto. BC3 Basque Centre for Climate Change, Bilbao.

Rockafellar, R.T., Uryasev, S. (2002), Conditional value-at-risk for general loss distributions. J. Bank. Finance 26, 1443–1471, http://dx.doi.org/10.1016/S0378-4266(02)00271-6

Watkiss, P., Hunt, A., Blyth, W. et al. (2015), The use of new economic decision support tools for adaptation assessment: A review of methods and applications, towards guidance on applicability. Climatic Change (2015) 132: 401, http://dx.doi.org/10.1007/s10584-014-1250-9

Zeng, S., Zhang, S. (2011), Real Options Literature Review. iBusiness 03, 43–48, http://dx.doi.org/10.4236/ib.2011.31007