Collapse of RCR Tomlinson and Program Portfolio Effect (PPE)

The industry and our profession was still trying to better understand the causes and lessons learnt from Lendlease’s $450m loss while the spectacular collapse of EPC contractor RCR Tomlinson after raising $100m in capital shocked everyone again! You might be interested to know that RCR has been operating for 120 years! So, how could they get it wrong?

Many believe that more collaborative “alliance-style” contracts will be the solution for these problems. I disagree, because while the type of the contract will definitely have impacts on establishing the platform for sharing risks between the parties reasonably, I truly believe in the absence of the sufficient and adequate risk engineering assessment, transferring the risks may not be the best way of risk management for maximum value for our investments. You can read my other articles on the good practice of risk allocation between Client and Contractor and also the Contingency X Factor for identifying the problem early at the links below:

https://www.dhirubhai.net/pulse/contractors-taking-too-much-risks-pedram-danesh-mand/

https://www.dhirubhai.net/pulse/could-risk-engineering-help-mr-mccann-lendlease-cost-danesh-mand/

The RCR's strategy of bidding aggressively for solar farms, undercutting competitors to win 13 of them and control 20% of the solar market in such a short period of time, has been blamed for its failure.

I have worked on a number of renewable projects (mostly solar and wind farms) and although these projects are not really technically difficult to design and build, they are quite hard to plan and make them commercially successful (from EPC’s perspective) – mainly due to supplier limitations, long lead times, high LD’s, full commissioning requirements, connection to grid, sign off by market operator, etc. Similar to power stations, EPC fixed price contracts have liquidated damages (LD's) that the contractor have to pay if the projects are not delivered on time. You get them a couple of months late and most of your tight margin is almost gone!!

In this article, I would like to briefly highlight another important aspect of risk and contingency management across project portfolio, i.e. Program Portfolio Effect (PPE), for EPC contractors, like RCR Tomlinson.

For many successful EPC contractors with a reasonable portfolio of projects, bidding on EPC projects with P50-P70 (i.e. 50% to 70% confidence level for schedule and cost success) is quite common in Australia. Let’s dig a bit deeper here. 

Modern Portfolio Theory (MPT) is a theory of finance which attempts to maximise expected portfolio return for a given amount of portfolio risk – or equivalently minimise risk for a given level of expected return – by carefully choosing the proportions of various assets.

Most organisations manage a number of projects concurrently. Hence, individual project confidence levels can roll up to higher or lower confidence levels at the program level not only depending on the project confidence level but also the level of correlation between projects. This is sometimes called the ‘Program Portfolio Effect’ – which is defined as the tendency for the risk on a well-diversified holding of investments to fall below the risk of most, and sometimes all, of its individual components.

For the Program Portfolio Effect to work, the budgeted contingency allowances should be managed at the program/portfolio level. This means the projects within the program that turn out not to require their entire original contingency budget must be managed in such a way that their unused contingency budget is available to other projects. By combining an active Program Portfolio Effect and having an efficient contingency management process in place, organisations can fund projects at lower confidence levels while achieving higher confidence levels from the organisation’s viewpoint.

Table below (AFCAA Handbook 2007) presents an example for analysis of portfolios sized with 5, 10, or 20 projects with high dispersion. The table shows assumptions of projects funded at probabilities of 50, 60, 70, and 80%. The third column shows the overall portfolio confidence level of each case with the projects uncorrelated.

Let’s test RCR’s case study (assuming 10 projects) against this. Considering their strategy of bidding aggressively for solar farms, we can assume (in the best case!) the Project Probability were at 60%. With 0.25% Correlation, this means 61% probability at their portfolio level. All good and reasonable?

NOT really! I have not worked or seen RCR bidding cost estimation, SRA or cost contingency models but even by using the most reasonable contingency determination methods, these probabilities are just indicative %’s, especially considering their experience in this new market. Now, quite interestingly if they got it just 10% wrong, i.e. Project Probability of 50% (which based on my experience on these projects I can definitely assure you it’s very easy to get it wrong that much!), the Portfolio Probability will drop to 36% from 61%! This example indicates how modest increases in each project’s cost-risk exposure (i.e. ‘shaving’ risk dollars by reducing each project’s probability) can lead to a significant reduction of a portfolio’s probability of meeting its funding level.

In other words, I may conclude that it was a possibility that they were running their projects at about 36% confidence level across the portfolio! WOW! So, if you think the recent raised $100m might be enough, think again! I can't make any further comment or forecast!

AND this PPE assessment is as important & critical for CLIENTS as for CONTRACTORS!! Setting your contractor for failure and risk passed to them will get back to you sooner than you think, like a boomerang!

Now, you can see why I truly believe in the absence of the sufficient and adequate risk engineering assessment from projects to portfolio, simply transferring the risks may not be the best way of risk management for maximum value for our investments.

Let’s finish with a couple of good news. The coming 2nd Edition of Risk Engineering Society (RES) Contingency Guideline, which will be published for public consultation soon, does have a section on this important topic.

We will also discuss about this topic and more at the next Risk Engineering and Project Controls Conference, 15-17 May, 2019, at ICC, Sydney. The registration is now open. Please come, let’s discuss and take accountability of our risk engineering and project control. We can – and must – help our projects and organisations not facing these challenges.   

Your views and comments are welcomed, as always.

The views expressed in this article are those of Pedram Danesh-Mand and do not reflect or represent the official policy, position or recommendation of the KPMG Australia, Engineers Australia (EA) or Risk Engineering Society (RES).