Editor’s note: Mr. Walker died on April 15, but a few of his previously submitted commentaries were as yet unpublished. This is the first part of one of them. Part two will follow next week.
The current airport expansion proposal reminds me of three definitions.
1. Camel: a horse designed by a committee.
2. Humped camel: a horse designed by two committees.
3. Elephant: a mouse designed by the Virgin Islands government.
While we are on the topic, one should be aware of some of the “ins and outs” of the contracting world.
1. It is a highly competitive world. Of the three “shortlisted” companies, two are going to find themselves out in the cold with nothing but bills to show for their efforts to get the contract. So it really behooves all three of them to bid as low as they can and to make their bid sound like the VI will get as much as possible for as little (in cash) as possible. At this point in time, half-truths are quite possibly the order of the day. Second place in this game is nothing but a dead loss.
2. Any contractor probably has several levels of accuracy of bid estimate to use for submitting proposals. To some degree, it may depend on how badly they want (or need?) the work and to another degree it depends on whether they assess their chances of winning as good or bad. No point in wasting both time and funds if you think your chances are not good.
3. There are possibly three generic types of contract that may be specified by the prospective client. A short description of each follows.
• A lump sum/fixed price contract is often most favoured by governments that make the mistake of going into the bid process with a poorly defined scope of work, poorly defined specifications, and vague requirements. (All of which seem to be characteristic of the VI government. See also the new hospital.) This type of contract is possibly the easiest to evaluate, but the fixed price seldom survives the breaking of ground. It is also totally dependent on the quality of the opposing project managers — and the VI has a historically poor record. This contract type is potentially the most profitable for the contractor, but also the most risky and the most likely to wind up in either civil or bankruptcy court.
b) A cost-plus contract is the least risky for the contractor, and it can be very profitable if the client’s scope of work is poorly defined. It is also subject to danger from inept project supervision on either side, but mainly the client’s. This type of contract is probably the most expensive for the client in the long run. (Again, see the new hospital, bid at $64 million and now over $100 million.) Cost-plus contracts are hardest to evaluate because part of the evaluation is just where the contractor is trying to build in traps. Other variants include cost plus fixed fee; cost plus fixed percentage; guaranteed completion date with a penalty/bonus clause; and anything else that clever negotiators can think of.
c) A guaranteed maximum price contract with a split in savings behaves like a cost-plus job up to the guaranteed maximum price and like a fixed price job after that point. If the total job cost comes in under the guaranteed maximum, then the under-run is split between the contractor and the client at an agreed percentage, commonly 75/25 or 90/10, client to contractor. This type of contract requires good project management and cost control on both sides. It is not as common as the other two.
Estimate accuracy
Depending on how much the contractor is willing to invest in bidding the project, the accuracy of the estimate will vary. The following are rough descriptions of what may take place.
a) In an order-of-magnitude estimate, almost no engineering is done prior to bidding. No equipment (other than very highly specialised items) is inquired. This estimate is generally considered to be for budgetary purposes only, and accuracy tends to be about 25 percent. Contingency reflects the accuracy and effort at estimated price, plus 25 percent. This is sometimes called a “cigarette package estimate.”
b) For a working estimate, sufficient engineering is done to reasonably define major project details. Some inquiry work is done on material and equipment. This type of estimate is often used to further reduce the number of bidders or to eliminate those that are obviously not really interested. Accuracy is about 10-15 percent, and contingency is equal to accuracy. A working estimate can be used to negotiate the final price.
c) For a definitive estimate, essentially all basic engineering is done and all major equipment is inquired. Labour is priced out; similar jobs in the same area (same country) are carefully examined. Accuracy and contingency are both about five percent. Definitive estimates are not often negotiated since there is so little left on the table. These estimates require the most effort to bid, and cost the most in the bid phase. They may be the most difficult to evaluate by the client.
Leader’s claims
It is now perhaps time to look at what Deputy Premier Dr. Kedrick Pickering said in a recent address to the House of Assembly. But to make real sense of this, one needs a bit of history of the verbiage that has gone before. Some of it is perhaps less than consistent or totally clear.
The base part of the proposed airport expansion is the 2,500-foot extension of the runway to accommodate large jet aircraft such as the Boeing 757 and 767. These planes have a passenger capacity of 250 to 300 passengers, depending on seating configuration and fuel load. So any extension of the runway is pointless unless the ancillary passenger handling capabilities of the airport complex are also expanded (and staffed), and unless the aircraft fueling and maintenance facilities are also brought up to standard.
In addition, any extension of the runway must meet standards set by the regulator, Air Safety Support International. This probably also includes United States Federal Aviation Administration standards since most aircrafts expected to use the facility will be from US airlines.
Cost-benefit analysis
In view of the conditions of the Protocols for Effective Financial Management, there also must be a “robust” analysis of the cost/benefit picture. Presumably this analysis should be based on such facts as can be accurately determined rather than wishful thinking as seems to be the case at present. All of the approvals — for the engineering design, the financial viability and all financial aspects, such as the availability of funds — will take time. It is my opinion that the time allowed by Dr. Pickering is not adequate given both the nature of the requirements and the delays that are already occurring.
To assess whether any company can achieve Dr. Pickering’s schedule, one must attempt to guess what resources are likely to be available to the successful bidder. The only information available is really the requirement for an annual business volume of $100 million. For most major engineering construction companies, the “home office” portion of annual business volume is usually somewhere in the 10 percent range. So it might be assumed that the contract disposes of an engineering/management/administration annual payroll of $10 million.
For an average hourly rate of $35 per hour and an overhead and profit rate of 100 percent, that equates to about 150,000 man hours per year, or 12,500 man hours per month. Assuming that, for the purposes of preparing a competitive bid that your chance of getting is no better than one in three, you employ no more than 30 percent of your workforce, you have available to you 3,750 man hours per month.
Further assuming that the overall project is in the $150 million range (consistent with the interest allowance of the VI’s 2013 budget), and assuming that 10 percent of the total engineering effort for the project is required to produce an estimate whose accuracy is 10 percent, and finally assuming that total engineering amounts to 10 percent of project cost (or $15 million), it is possible to estimate how long it will take to prepare the figures that are required for the initial submission. In fact, this was March 13, but the submissions are not adequate for contract award. Thus, it is probable that the government now finds itself in a position where the cost is already too high and must (somehow) be reduced to a more reasonable figure that has some possibility of passing United Kingdom scrutiny.
Eight-month delay?
The result of this very approximate calculation is more than eight months! And this figure will apply at this point if the “successful” three are now required to produce a more accurate effort.
This can, of course, be at least partially countered by awarding the project on the basis of the
March 13 bid packages (even though they are known to be inaccurate) and paying the successful bidder to re-estimate the project based on more complete engineering. This, of course, requires a further delay while better data is developed and analysed. Since the contractor now has some assurance that he or she has the job, the level of effort can be much higher. Of course, there must remain, despite all of Dr. Pickering’s assurances, the nagging suspicion that the project will never go ahead. So, even if you are in the “catbird seat,” you may not wish to waste a lot of time and money.
And, in reality, the cost-benefit analysis should be based on the final estimated cost, not the earlier one. Or perhaps a “guaranteed maximum” contract with a 90/10 split in savings is the way to go?
To be continued in next week’s edition.
{fcomment}