Escalating to the start on site leaves out almost a million, escalating to completion adds one, and on a back-loaded scheme even the midpoint is short.
Escalate the base-date estimate to the midpoint of construction, compounding the forecast inflation rate over the months from the base date to that point: factor = (1 + rate)months ÷ 12. On an illustrative 24,000,000 estimate, nine months before a 24-month build at 4.0 per cent a year, that is 21 months and 1,705,117 of escalation, 7.10 per cent. Stopping at the tender date leaves out 988,658; running to completion adds 1,028,205 that will never be spent.
Worked in full in Construction Cost Control by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
Every cost plan is priced at a base date, and every building is paid for later, month by month, over the whole construction period. Escalation is the bridge between the two. The midpoint convention is a shortcut for the cash flow, and it works well when the spend is roughly symmetrical. Knowing when it stops working is the part that matters.
| Input | Value |
|---|---|
| Construction cost at base-date prices | 24,000,000 |
| Months from base date to start on site | 9 |
| Construction period, months | 24 |
| Forecast tender price inflation a year (illustrative) | 4.0% |
| Months from base date to midpoint of construction | 21 |
The inflation rate should be a tender price forecast, not a general price index: tender prices carry contractors' margins and market appetite and they move differently from materials and labour costs. In the UK the usual source is a published tender price index forecast. The 4.0 per cent here is illustrative.
Months to midpoint = months to start + construction months ÷ 2 = 9 + 12 = 21
Escalation factor = 1.0421 ÷ 12 = 1.0710
Escalation = 24,000,000 × (1.0710 − 1) = 1,705,117
In Excel: =Cost*((1+Rate)^((Start+Build/2)/12)-1). Keep the rate and the dates as inputs on the cost plan summary, so that a later start date reprices the plan on its own.
| Escalated to | Months | Factor | Escalation |
|---|---|---|---|
| Start on site only | 9 | 1.0299 | 716,459 |
| Midpoint of construction | 21 | 1.0710 | 1,705,117 |
| Completion | 33 | 1.1139 | 2,733,322 |
Escalating to the start date only is the most common error in early cost plans. It treats the building as if it were bought in one payment on the first day on site. The 988,658 it leaves out is real money: on a fixed-price contract the contractor prices it into the tender, and on a fluctuating contract the client pays it through the formula. Either way the client pays it.
Escalating to completion makes the opposite error. It inflates the first month's concrete as though it were bought on handover day, and adds 1,028,205 of contingency disguised as escalation. Simple interest instead of compound gets close, 1,680,000 against 1,705,117, and is a small error, 25,117 here, that grows with the period and the rate.
The midpoint is a stand-in for the centroid of the spend: the date by which, on average, each pound is paid. On a symmetrical S-curve the two coincide. A cash flow escalated month by month on a symmetrical curve comes out at 1,709,085, only 3,968 above the midpoint figure.
Many buildings do not spend symmetrically. Residential and fit-out heavy schemes back-load: finishes, services and fit-out are a large share of the cost and they come late. On an illustrative back-loaded curve with 62.1 per cent of the spend in the second half of the programme, the centroid sits at month 13.6 of the build, 22.6 months from the base date, and the month-by-month escalation is 1,841,007: 135,890 more than the midpoint shortcut.
| Spend profile | Centroid, build month | Escalation | Versus midpoint |
|---|---|---|---|
| Midpoint shortcut | 12.0 | 1,705,117 | 0 |
| Symmetrical S-curve, month by month | 12.0 | 1,709,085 | 3,968 |
| Back-loaded S-curve, month by month | 13.6 | 1,841,007 | 135,890 |
If a cost plan is cash-flowed anyway, as it must be for the development appraisal's finance line, escalate each month's spend from the base date to its own month and sum. The midpoint is for the stage before a cash flow exists.
| Inflation a year | 18-month build | 24-month build | 30-month build |
|---|---|---|---|
| 2.0% | 723,588 | 846,290 | 969,600 |
| 4.0% | 1,454,305 | 1,705,117 | 1,958,400 |
| 6.0% | 2,192,083 | 2,576,422 | 2,966,400 |
The rate is the input people argue about, and it moves the answer most. The date moves it too, and is argued about less. Push the start on site back six months at 4.0 per cent and escalation rises from 1,705,117 to 2,214,179: 509,062, or 84,844 for each month the design or the planning consent slips. That is a cost of delay that sits in no programme and no claim, because the client carries it before anyone has signed anything.
The common mistake is to treat escalation as a fixed allowance set once, at the first cost plan, and carried forward unchanged while the start date moves. The figure was correct for a date that no longer exists. Tie it to the programme with a formula so that it moves when the dates do, and report it as its own line rather than folding it into contingency, where it disappears and contingency appears larger than it is.
The second mistake is to count it twice. A contractor's tender on a fixed price already includes its own escalation to the dates it expects to buy each package. A client cost plan that compares itself with that tender must be escalated to the same dates, or it will show a false overrun; and a client cost plan that adds escalation on top of tendered prices counts it twice.
Escalation is one of the inputs that move a final account away from its contract sum, and one of the few that exists before signature. The free companion workbooks for this case run it as a one-input stress alongside design completeness, provisional sums and time. On the developer's side, the finance cost of the same programme is worked in how to calculate interest during construction.
Because the building is paid for month by month over the construction period, and on a roughly symmetrical spend curve the average payment falls at the midpoint. On an illustrative 24-month build, a month-by-month escalation of a symmetrical cash flow gives 1,709,085 against 1,705,117 at the midpoint, a difference of 3,968 on 24,000,000.
Compound, because each year's inflation applies to prices already raised by the year before. Over 21 months at 4.0 per cent on an illustrative 24,000,000, compounding gives 1,705,117 and simple interest 1,680,000. The error is small at low rates and short periods, and grows with both.
Every month of slippage moves the midpoint a month later. On an illustrative 24,000,000 estimate at 4.0 per cent a year, a six-month later start raises escalation from 1,705,117 to 2,214,179, or 84,844 a month, and the client carries it before any contract is signed.
Chapter 14 of Construction Cost Control runs escalation as one of six one-input stresses on the final account; the free Sensitivities workbook prints every line of each scenario. The book takes the same case from first principles to the decision, chapter by chapter, and every figure it prints is a live formula in the free companion workbooks.
Get the book on Amazon →Free companion files
Also on Amazon UK · Amazon Germany · Amazon France · Amazon Canada
Reading guide: business operations: cost, contracts, pricing and supply chain → · All 453 articles →
If this book helped, or didn’t, a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.