Analysis of “Shipbuilding contract holds $250M mystery” by Terry Milewski

Mr. Milewski’s underlying assumption in the article is stated in his byline: “Cost of Arctic patrol ships' design sparks warning of another procurement ‘fiasco.’ ”  The facts offered in support of the article by the CBC include:

  • Ottawa will pay Irving Shipbuilding $288 million just to design — not build — a fleet of new Arctic offshore patrol ships.
  • CBC surveyed similar patrol ships bought by other countries and they paid a fraction of that $288 million to actually build the ships — and paid less than a tenth as much for the design.
  • The design of Canada’s new ships is based upon a Norwegian vessel whose design Ottawa has already bought for just $5 million.
  • The Norwegian ship, Svalbard, was designed and built for less than $100 million in 2002.
  • Denmark acquired two patrol ships for $105 million in 2007 with similar ice-breaking capability.
  • Minister Peter MacKay explained that the costs from other shipyards are wrong.
  • An unnamed senior shipbuilding source said that the size of the Irving design contract is “inexplicable.”
  • The same unnamed senior shipbuilding source stated that the Svalbard design would need extensive revisions to adapt it to Canadian purposes and produce final blueprints, driving the bill up to $20 million.
  • Minister Rona Ambrose also insisted that spending more on the design would save money later on.
  • Much of the design work is actually going overseas.

Instead of assuming that the project is a $3.1 billion boondoggle, we work from the assumption that the stakeholders are doing their best, because they know that the risk is very high that cost overruns would create adverse public opinion and create a problem for everyone with an interest in the success of the project. The CFPS Maritime Security Analysis Team analysis follows.

The ‘apples versus oranges’ analogy used by several commentators stems from the fact that there is no common method to report costs from country to country. Some countries report just the cost of a construction contract while others, like Canada, fold much of the project and life cycle spares costs into the budget. The construction costs for the Danish patrol ship is a good case in point.

The Danish price reported by the CBC represent only construction costs in Poland for their vessel. However, cost has a relationship primarily to weight, not length as quoted by Mr. Milewski. The current estimated displacement of the AOPS is 5,730 tonnes while the Danish Knud Rasmussen-class patrol ship displaces 1,720 tonnes, or just 30% of the size. Using Danish numbers, an AOPS-sized vessel built in Poland would cost $172 million. Six would cost upwards of $1.05 billion and $1.4B for eight. Added to this must be non-construction project costs, which the Danes paid separately.

The cost figures for the Norwegian Svalbard date from 2000 and need qualification as well: is this just construction cost or the equivalent of the total Canadian project cost? It is unlikely to be the to total cost. Consider this: the cost to build a basic tanker in Korea is at least $100 million today. The design complexity of even a simple warship is far greater than a merchant ship. If the figure quoted for Svalbard is construction cost alone then, as a rule of thumb, the cost to build a ship represents only about a quarter of the total cost of a Canadian shipbuilding project. That would mean that acquiring a US $100 million Svalbard as part of a six ship Canadian project and factoring 10 years of 2% inflation (~33%) could be as much as 6 x 4 x (100 + 100/3) = US$3.192 billion.

The CBC report also implies we will be paying a significant premium for building the AOPS in Canada and that other nations build ships for much less. This is another argument unsupported by facts. There are but two studies we are aware of that looked into this ‘Canadian-built cost premium’ in any detail. DND’s audit arm issued a report that concluded that the built-in-Canada Canadian Patrol Frigate ultimately cost only 7% more (roughly $28 million per ship) on average than seven other similarly sized foreign warships. That 7% “at home” premium also created over 7,000 person-years of Canadian employment and established at least 12 Canadian companies that are still in business and exporting complex marine systems to such demanding customers as the United States, Israeli and Royal Navy today. (Read more about this here.) That same audit also considered the Canadian frigate the combat superior of every one of the contemporary foreign frigates studied save one to which it was the combat equal. The audit also quoted Forecast International, a US publication that conducts an annual assessment of warship capability, which concluded:

After a very shaky start, mainly due to the long gap in Canadian warship construction, the Halifax-class frigates have matured into fine warships. The lead ship of the class has been the subject of unstinting praise from the US Navy, following visits to American naval bases.  HMCS Halifax is also regarded as being a very satisfactory and a well-conceived design by the British Royal Navy Directorate of Navy Construction.

More recently, a 2009 study commissioned by Industry Canada from the engineering consultancy group Mott MacDonald, questioned similar doubtful claims of cheaper foreign shipyards and estimated that their products normally resulted in a 25% increase in in-service support costs after they were delivered. (Mott MacDonald, Economic Analysis of National Shipbuilding Procurement Practices, Industry Canada, 2009)  These costs, by the way, are not insignificant and can easily be double the cost to just build a ship. In addition, foreign firms have also been front and center recently in their readiness to demand more than was budgeted during both the earlier and current Joint Support Ship projects while also disputing elements of our procurement processes.

The CBC report uses the term “design cost” as if there was only one cost for the design.  Nothing could be further from the truth. Typically, the cost of design progresses in ever increasing detail and amount from the concept, to feasibility, to preliminary (where we are now), and finally to the definition stage. The Norwegian design that Ottawa  “already bought for just $5 million” represents a starting point for Canadian design work that culminates with the contract signed with Irving takes the design effort to the definition level.  The intention was never to build Svalbard-class ships in Canada. The Norwegian ship was only a conceptual starting point for a ship more suited to Canadian environmental conditions and operational requirements. This point is actually very important.

The approach by the AOPS project to sign this contract with Irving Shipbuilding likely represents a prudent, low-risk strategy to develop the Norwegian basis of a concept design to a level that will permit a very accurate cost estimate. There are considerable risks involved in a first-of-class Canadian design and construction project. Before Irving bids on the construction, all the production drawings and material requirements will be known and a test section will be built. This is necessary to set up the project for success, as there is no prior Canadian construction history for this class of ship from which to draw accurate analogies. The Danes and Norwegians have a great deal of experience in building earlier vessels of this type. The quote by Minister Ambrose that spending more on the design would save money later could be read as a politician’s way of saying that this contracting approach offers lower financial risk for the project. We support this view.

This definition contract, which is incorrectly called a design contract by CBC, is actually old news; it was announced on March 7th. While design is part of the total contract price, the definition contract is a task-based contract divided into seven work packages (or tasks) that could be awarded. C$135 million is tied up in the first two packages. A large part of the remainder is to pay for access to production of major components required for construction. Another major component is $38M for HST.

Will all tasks in the definition contract be awarded and when? These are the prudent questions for an investigative report interested in truth. The CBC report does not ask how the Danish designer fits into this contract; it is likely only for one of the seven tasks. Equating the total potential value of the contract to the amount paid to a Danish naval architecture firm is a very, very large leap in logic, and an error.

This contract will allow Irving Shipbuilding to refine and complete the AOPS design to the production level prior to the commencement of construction in 2015. This will permit the estimation of construction costs with far great accuracy than building directly from plans. Considering that many subcontractors are involved, this is much more than just “design,” it includes production planning, design integration and materiel sourcing. This cautious approach to contracting in a progressive manner reduces risk, the chance of huge cost overruns in the project, and it offers the government off ramps if the project goes ‘pear-shaped’ at any point before construction.

This definition contract will be followed by a construction contract in 2015. Under the contract Irving Shipbuilding have the next 30 months to:

  • Refine and complete the AOPS design to the production level, including 3D designs;
  • Choose specialist subcontractors and material suppliers;
  • Complete various phases of design review, analysis and testing;
  • Plan how to integrate all of the complex systems and equipment on the ship;
  • Do detailed cost estimates for the build phase of the project;
  • Complete and test a production module to verify design, engineering and systems integration plans;
  • Employ an estimated 110 additional workers, bringing the Irving Shipbuilding AOPS team to 200 [the wages alone for 110 people employed at the shipyard for 30 months at an average salary and benefit package of $80k could run to $22 million].

This is not a design contract; it is a definition contract and the work involved is often part of a build contract. The proof of success will be whether the construction contract will include six or more vessels and still keep the project within the budgetary ceiling. The work in this first contract will determine if this is possible. The signing of a construction contract to cut steel and begin building in 2015 will depend on the outcome.


Leave a comment

Your email address will not be published. Required fields are marked *