A buying group runs a tender for work gloves on behalf of its members. The existing framework agreement has an annual volume of 400,000 euros. Five suppliers are invited and submit a bid. The lowest first bid comes in at 440,000 euros, the average of all bids at 480,000 euros, the highest at 560,000 euros. After negotiation the lowest bidder wins the contract at 372,000 euros.
The numbers are constructed but realistic. The question is which saving the central office now reports to its members.
Measured against the existing framework agreement it is 7.0 percent. Against the winner's first bid, 15.5 percent. Against the average of all bids, 22.5 percent. Against the highest bid, 33.6 percent.
All four values are arithmetically correct and can be stated in a report. They describe the same procedure. Setting aside a question about the reference basis, each of them survives scrutiny.
That is the point. None of the values is wrong. They are simply not the same thing.
Only the first is economically relevant.
The difference of 28,000 euros against the existing framework agreement changes next year's result. It is cash-effective. It shows up in the profit and loss accounts of the member companies that buy these gloves.
The difference of 188,000 euros against the highest bid changes nothing, because that price never stood to be paid. Nobody ever intended to pay 560,000 euros. The amount describes how far the bids were apart, and that is a statement about competitive intensity, not about profit.
Between the two lies a factor of nearly seven. Same tender, same suppliers, same day.
In a buying group the benefit arises at the member company while the report about it comes from the central office. That separation makes the question of the reference basis more important than it is in a single company.
If the central office reports 33.6 percent and the member company finds none of it in its own accounts, distrust arises that carries over into the next joint tender. If it reports 7.0 percent and explains what the calculation was made against, the number is smaller and the statement holds.
The reflex points the other way, because a bigger number looks like more performance. In an organization whose members can follow the calculation themselves and will check it if in doubt, the smaller number is the better route.
There is nothing wrong with reporting both. Cash-effective saving and avoidance effect separately, each with the stated basis. What does not work is addition. Adding cash-effective and non-cash-effective figures produces a total that is neither one nor the other.
When a category is tendered jointly for the first time, no reference price exists for the group. By definition there is then no cash-effective saving, regardless of how well the tender went.
In buying groups this is not the exception but the normal case at the start. Three statements nevertheless remain possible. The gap between the winner's first bid and the awarded price describes the concession of a single bidder. The gap to the median of all bids describes the value of competitive intensity. The gap to the estimated price describes the quality of the estimate.
Of these three, the gap to the median is the most informative, because it answers the question of what the involvement of several suppliers contributed. We treat it as an avoidance effect and report it separately from cash-effective amounts.
Which reference basis is the right one? The price last actually paid, at the same volume and the same specification. Where the contract would have been renewed without the tender, it is the valid framework agreement price. In volatile categories it is the previous year's price adjusted for raw material and index movements, and that is also the most laborious variant.
Are we allowed to report the avoidance effect at all? Yes. It has its place, especially in first-time tenders. It simply has to be labelled as what it is, and it must not sit in one total with cash-effective amounts.
How do we check a figure from a live tender? With one question. Did the price being calculated against ever stand to be paid? If not, the number describes avoided expense, not a saving.
Chapter 2 of our study contains this example calculation in full, together with a taxonomy of the six common reference bases and the five adjustments without which a price comparison is not valid.
→ To the white paper
The numbers are constructed but realistic. The question is which saving the central office now reports to its members.
Four numbers that are all correct
Measured against the existing framework agreement it is 7.0 percent. Against the winner's first bid, 15.5 percent. Against the average of all bids, 22.5 percent. Against the highest bid, 33.6 percent.
All four values are arithmetically correct and can be stated in a report. They describe the same procedure. Setting aside a question about the reference basis, each of them survives scrutiny.
That is the point. None of the values is wrong. They are simply not the same thing.
Which one reaches the members' accounts
Only the first is economically relevant.
The difference of 28,000 euros against the existing framework agreement changes next year's result. It is cash-effective. It shows up in the profit and loss accounts of the member companies that buy these gloves.
The difference of 188,000 euros against the highest bid changes nothing, because that price never stood to be paid. Nobody ever intended to pay 560,000 euros. The amount describes how far the bids were apart, and that is a statement about competitive intensity, not about profit.
Between the two lies a factor of nearly seven. Same tender, same suppliers, same day.
What this means for the central office's success report
In a buying group the benefit arises at the member company while the report about it comes from the central office. That separation makes the question of the reference basis more important than it is in a single company.
If the central office reports 33.6 percent and the member company finds none of it in its own accounts, distrust arises that carries over into the next joint tender. If it reports 7.0 percent and explains what the calculation was made against, the number is smaller and the statement holds.
The reflex points the other way, because a bigger number looks like more performance. In an organization whose members can follow the calculation themselves and will check it if in doubt, the smaller number is the better route.
There is nothing wrong with reporting both. Cash-effective saving and avoidance effect separately, each with the stated basis. What does not work is addition. Adding cash-effective and non-cash-effective figures produces a total that is neither one nor the other.
The special case buying groups often face
When a category is tendered jointly for the first time, no reference price exists for the group. By definition there is then no cash-effective saving, regardless of how well the tender went.
In buying groups this is not the exception but the normal case at the start. Three statements nevertheless remain possible. The gap between the winner's first bid and the awarded price describes the concession of a single bidder. The gap to the median of all bids describes the value of competitive intensity. The gap to the estimated price describes the quality of the estimate.
Of these three, the gap to the median is the most informative, because it answers the question of what the involvement of several suppliers contributed. We treat it as an avoidance effect and report it separately from cash-effective amounts.
Frequently asked questions
Which reference basis is the right one? The price last actually paid, at the same volume and the same specification. Where the contract would have been renewed without the tender, it is the valid framework agreement price. In volatile categories it is the previous year's price adjusted for raw material and index movements, and that is also the most laborious variant.
Are we allowed to report the avoidance effect at all? Yes. It has its place, especially in first-time tenders. It simply has to be labelled as what it is, and it must not sit in one total with cash-effective amounts.
How do we check a figure from a live tender? With one question. Did the price being calculated against ever stand to be paid? If not, the number describes avoided expense, not a saving.
The next step
Chapter 2 of our study contains this example calculation in full, together with a taxonomy of the six common reference bases and the five adjustments without which a price comparison is not valid.
→ To the white paper