Managing Assets/Projects using Portfolio Strategy

July 18, 2011 • Tell FriendsPrinter Friendly

This article discuss how we can apply portfolio theory in managing our assets or project to achieve a company’s goals and targets.


A Company should routinely apply portfolio management techniques to optimize the value of their assets and projects. The portfolio methods were used to give valuable quantitative analysis for management to forecast and monitor a company’s performance against specific goals and targets on a probabilistic basis.

The concept of portfolio is theoretically well-established and has been extensively applied during the past 50 years. The method integrates financial, risk, and statistical analysis techniques. The application of portfolio methods includes portfolio theory, Monte Carlo simulation, cost and productivity of capital using discounting techniques, and expected monetary value (EMV). This method enables analysts and the decision-makers to rank different portfolios of projects in terms of the probability of achieving the strategic targets specified by the company.

Quantifying risk is central to modern portfolio methods. The risk is measured using the mean deviation of all value outcomes from a Monte Carlo simulation that fall below the target value. This measure extend portfolio theory, which as originally applied to the financial sector sought to minimize risk by focusing on the lowest standard deviations of portfolio value distributions.

EMV (not NPV) of the portfolio should be used to define the efficient frontier as it also incorporates the detailed risk analysis (i.e., chance of success) associated with the individual projects. By ignoring the geological, engineering, technological, and political risks, the efficient frontier defined using NPV and downside (economic) “risk” will usually present a misleadingly optimistic view of a portfolio’s value and downside risk.

There is also value in identifying economic risks that are independent of the corporate financial structure and comparing them with risks associated with achieving the corporate targets.
A portfolio-monitoring process is therefore proposed here that analyzes risk in three stages, as described in below chart. This approach was developed by David Wood (2008)

Chart 1.
Three Stages of Risk Method (Wood, 2008)


A Company portfolio of three exploration, five production, and two development assets is used in this study to illustrate how a model built on the above design can aid portfolio management.

Table I. Summary Description of Company Portfolio

Table 1 lists key production, reserves, costs, and asset risk factors for each asset. Each asset has its own distinctive cost, production, and reserves profiles developed over the 24-year period year 1-28. On an un-risked basis total reserves for the portfolio amount to 372 million bbl of oil and 3,414 million btu of gas, whereas on a risked basis these reserves fall to 83% of oil and 56% of gas.

Table 2. Economic Input Variables to Simulation Model

Table 2 lists the input variable distributions used for the economic models. Six variables are considered as triangular distributions. The most likely value of each distribution is taken as the likeliest and used for the deterministic evaluation. All six variables are considered as dependent distributions for all projects. In each of the 1,000 iterations used for the Monte Carlo simulation one random value from these dependent distributions is applied to each project.

Oil price was assumed in year 1 at $32.0/bbl and would be declined up to year 5 with 5 % discounted until the equilibrium price achieved beyond year 5 at $25.00/bbl.

The rest of input variable represent the performance of target achieved and spending capability in production, development and exploration activities.

Table 3. Before Tax, Risked and Unrisked Cash flow Values calculated by Simulation and Deterministically

Table 3 lists the total NPV, total EMV, and ratio yardsticks calculated for each block based upon the deterministic calculation of the most likely case (i.e., mode) and the mean of the simulation analysis distributions.

For the total portfolio the mean total EMV is about 100.33% of the mode total EMV. The higher mean value calculated by the simulation means the input variable distributions had been skewed to the upside values. The mean total EMV for the portfolio is about 33% of the mean total NPV, illustrating the impact of taking into account inherent project risk.

A ranking of the projects in value order is significantly different if based upon NPV rather than EMV criteria. Also, a ranking based on total EMV/I is slightly different from that based on total EMV (e.g., compare PROD 2 and 3 block), which is a consequence of the relative capital requirements and timing of capital expenditure.

In the production blocks, PROD 2 has a lowest EMV/I. It seems the capital requirement on this field is the biggest one compares the EMV generated.
In the development blocks, DEV 2 is better than DEV 1 in EMV/I ranking even though the chance of success is lower.
In the exploration blocks, EXPL 3 is the lowest EMV/I. It seems this block is the first priority to be farmed out to share the risk in this block.

Analyzing the Company Assets on Production Sharing Contract (PSC) Basis

For each project, six calculated parameters are recorded for each year for each iteration of the simulation. The parameters recorded are real EMV, real NPV, undiscounted and unrisked cash flow, risked and discounted real capital expenditures (capex), risked annual production, and risked remaining reserves (at start of year). The EMVs and NPVs are calculated on a PSC basis at this stage. The data are then summed to give portfolio distributions for each parameter (i.e., 1,000 data points for each year). These distributions are then analyzed statistically.

Fig 1. Cash Flow for Company Portfolio

Fig. 1 illustrates some of the statistically defined profiles for the EMV distributions of the portfolio and compares them with the deterministic mode values. The lowest risked and discounted cash flow (EMV) occurs in year 4 (mean ~ -$53.4 million; mode ~ -$53.3 million), EMV peaks in year 6 (mean ~$53.3 million; mode~$53.1 million) and then declines progressively to year 24 (mean ~ $6 thousand; mode ~ $6 thousand).

Fig 2. Cumulative Probability Distribution of total expected monetary value for portfolio

Fig. 2 shows the portfolio’s cumulative probability distribution for ?EMV with P10 (i.e., 10% chance of being less than) and P90 values of $117 million and $188 million respectively. The mean of the total EMV distribution is $151.9 million, which is very close to the P50 value of $151.2 million. The mode or most likely deterministic total EMV of $151.4 million is similar to the central tendency values of the calculated probabilistic distribution.

Fig 3. Annual Capital Investment for Company Portfolio

The high downside risk is associated with the early years when the major capital investments are made (Fig. 3). The downside risks associated with years beyond year 6 are all less than zero.

Defining the PSC Efficiency Frontier

For the 10 existing projects in the portfolio it is informative to selectively remove all or parts of one or more of the blocks and re-sum the 1,000-iteration total EMV data points in order to calculate total EMV and mean downside risk for the remaining group of projects. The calculations was used to define a provisional efficient frontier (PSC base). Sorting and ranking the data in descending order of total EMV and downside risk can identify the highest value combination of projects for given levels of risk.

Fig 4. Efficient Frontier Based Upon PSC total EMV’s and A Measure of Downside Risk

Fig. 4 plots total EMV against semi-standard deviation for portfolios from which one or two projects are removed. The downside risk measurements show symmetrical relationships. There is a significant spread of risk for portfolio combinations with total EMVs greater than $150 million.

Table 4. Ten Company Portfolio Combination Ranked by ?EMV

Table 4 lists the top 10 total EMV ranking portfolios identifying those blocks excluded, their downside risk values, and associated risked and discounted total capex. The portfolio with all 10 blocks has the highest total EMV of $225 million, semi standard deviation of $11 million and total capex of $538.

Table 4 shows that the portfolio excluding EXPL 3 block has its value added by about $54 million for a reduction in semi standar deviation about $1.3 million (i.e., a 10% reduction in downside risk). Excluding both of the lowest value projects (i.e., EXPL 2 and 3; see Table 3) from the portfolio adds value and reduces downside risk by $73 million and $2 million, respectively. The small contribution of these two projects to portfolio value should lead to a careful study of what the projects are costing in terms of management time and general overhead. Corporate strategy may indicate that these are candidates for sale or farm-out.

It is apparent from Table 4 that the portfolio excluding EXPL 2 and 3 blocks leads to the lowest downside risk of $11 million (i.e., a 15% reduction in downside risk relative to the full portfolio) for a addition in portfolio value of $73 million. The reduction in risk is clearly correlated with the amount of capex associated with each portfolio. EXPL 2 and 3 require the highest and second highest amounts of ?capex in exploration blocks, respectively, of all the projects in the portfolio but have only the tenth highest and ninth highest total EMV respectively (Table 3). Excluding these two projects reduces risked capex from $626 million to $538 million (a reduction of 14%).
If capital is tight or the corporate strategy is to minimize risk then it may be decided to remove (i.e., sell, farm out, or not pursue) these blocks from the portfolio. Of course, in a real portfolio there may be contractual obligations that prevent such blocks being terminated or delayed. The high individual contributions of these two projects to downside risk are not due primarily to the inherent project risks (EXPL 2 and 3 is an exploration block with a 15% chance of success) but rather to the high capex and low investment efficiency of those projects.

Fig 5. Efficient Frontier Based Upon PSC total EMV / I ‘s and A Measure of Downside Risk

* referred to table 4 for the symbol of the numbers

Fig. 5 plots the efficient frontier using total EMV/I (i.e., risked investment efficiency yardstick) rather than the risked value ?EMV. The higher investment efficiencies of those portfolios with reduced downside risk on the efficient frontier constitute a clear indication that downside risk is strongly influenced by ?capex in this portfolio.

This analysis was taken further by analyzing the effect of including fractions of projects and/or combinations of parts of EXPL 1, 2, and 3 blocks in the portfolio. This is equivalent to a joint venture or farmout arrangement with no promote being paid by the farminee. Promoted joint venture terms could also be included in such analysis for a real portfolio if such terms are realistically attainable. Fig. 5 illustrates the results of such an analysis.

Fig 6. Efficient Frontier Based Upon PSC SEMV’s
and fractions of selected project combinations

* referred to table 4 for the symbol of the numbers

Fig. 6 shows the effects on value and risk of fractions of EXPL 3 block combined with the remainder of the portfolio. Comparing Farm out 50% (portfolio 5) , 75% (portfolio 6) and 100% (portfolio 1) of EXPL 3 Block results in a significant addition in value for high variation in risk specially for 5 and 6 portfolios.

Including fractions of EXPL 2 (portfolio 7) and EXPL 1 (portfolio 8) besides EXPL 3 results in a significant reduction in risk for little variation in value (i.e., near horisontal but non-linear trends in Fig. 6). These are key projects for the portfolio, and removing them even in part dramatically adds overall portfolio value. On the other hand, including fractions of projects EXPL 1, 2, and 3 results in a significant reduction in risk for high addition in value (i.e., slightly inclined but non-linear trends in Fig. 6). The projects represent the more marginal contributers to portfolio value.

Analyzing the Company Portfolio of Projects on a GAAP Basis

Net income and GAAP cashflow (referred to here as GAAP EMV) are the two calculated financial metric distributions from the simulation that reveal most about the value of the portfolio to Company and potential company performance problems that may lie ahead. Net income and GAAP EMV are recorded for each year on an undiscounted basis and discounted at the corporate rate (i.e., 15%). Both discounted and undiscounted distributions are worthy of analysis, but it is the discounted distributions that are illustrated here in Fig. 7.

Fig 7. Net Income and GAAP EMV distributions for Company portfolio based Upon GAAP

As should be expected the statistics for the annual distributions show quite different trends for net income and GAAP EMV. Net income for the portfolio shows increased discounted values for years year 1 to 6 (mean values between $38 million and $100 million) which gradually decline to less than a mean of $5 million beyond year 7. In year 23, net income dropped -$10 million because we assumed PROD 2 as the biggest asset would be expired, so the book value would be zero in this year.

In contrast, GAAP EMV shows negative values in the first 4 years with a clear trough in year year 4 associated with the capital expenditure program. GAAP EMV rises to a peak in year 7 (mean values of $52 million) and then declines to a mean value of less than $2 million for years beyond 23. The probability distributions for both distribution sets show more uncertainty for the first quarter of the period modeled than for later years.

It is particularly instructive for short and medium term planning to relate these trends to annual corporate targets. In Fig. 7 some hypothetical corporate targets are shown for each parameter. For net income the target is $60 million in year 1 escalating in real terms at 6%/year. For GAAP EMV the target is $10 million in year 1 escalating in real terms at 6%/year. These targets are discounted at 15%/year to correspond with the discount rate for the calculated values. For each year the probability of achieving these targets is calculated.

With respect to its targets, Company clearly has some problems with its existing portfolio of 10 E&P blocks.

Key problems to solve are:

  1. High capital expenditure in year 1 to 4 (Fig. 3) required to prove up reserves and maintain production presents a short-term negative cash flows.
  2. The chance of achieving corporate net income targets is only about zero for the first 4 years. This needs to be improved if possible.
  3. The performance targets need to be refined for specific periods. However, even lowering the targets significantly does not yield a realistic chance of achieving an acceptable net income beyond about year 7 or acceptable cash flow beyond year 10.

Figs. 7 highlight the need to set a planning horizon that can address meaningful corporate targets. Problems 1 and 2 above are short to medium term that need to be addressed in the very near future if the corporate targets for year 1 to year 5 are to be met. On the other hand, Problem 3 is a long term problem that cannot be dismissed but needs to be the focus of longer term strategy. Problem 3 is the result of short-life reserves in the portfolio with no new reserves added beyond year year 5 from the 10 blocks.

The company lacks a long-life income stream such as from infrastructure or processing tariffs. It is the nature of the E&P business and related natural resource production industries that longer-term cashflow and earnings from proved reserves currently being produced show quite steep declines. It is for this reason that E&P companies have new business divisions to add new reserves in the longer term through both exploration and acquisitions.


This analysis leads to several useful conclusions:

  • During year 1 – 4 period, there is a significat risk of negative EMV due to major capital investment program in DEV 1 and 2 blocks. The lowest risked and discounted cash flow (EMV) occurs in year 4 (mean ~ -$53.4 million), EMV peaks in year 6 (mean ~$53.3 million) and then declines progressively to year 24 (mean ~ $6 thousand; mode ~ $6 thousand).
  • From the various scenario portfolio as shown in the efficient frontier graph, the portfolio excluding EXPL 3, EXPL 2 and EXPL 1 leads to add the EMV and reduce the downside risk. It may be indicated that these blocks are candidates for farm out some to share the risk.
  • The high individual contributions of these three exploration blocks to the downside risk are not due primarily to the inherent project risks (exploration blocks was assumed 15% chance of success) but rather to the high ?capex and low investment efficiency of those projects.

Subscribe to Real Options Valuation by email to get course schedule and updates from us.