Cash flow, budget and cost control plan

Ruth looked at Tomas's first cash flow chart for a long time before she said anything. 'So we will be about two and a quarter million dollars out of pocket in month seven,' she said. 'Who is paying for that?' Tomas had not thought of it that way. The estimate said Tamarack would make a fee of about $720,000. The cash flow said that, for most of the job, Tamarack's bank account would be well below zero. Both were true. A contractor that understands only the first number can win profitable jobs and still run out of cash.

This lesson covers the seventh deliverable: the cost-loaded schedule, the project cash flow, and the plan for controlling cost once construction starts. It draws on project cost control and earned value (CON-310) and construction finance and accounting (CON-320), and joins the estimate from lesson 5 to the schedule from lesson 6.

From estimate to control budget

An estimate is organized to win a bid. A control budget is organized to run a job. After award, the project team turns the estimate into a budget by cost code, so that every invoice, timesheet and subcontract payment can be charged to a code and compared with the money allowed for it. Tamarack's cost codes combined the WBS area from lesson 2 with a cost type: labour, material, equipment, subcontract or general conditions. The fee was held separately and not spent.

The first months after award are the buyout period, when the project manager signs subcontracts and purchase orders for the packages priced on bid day. Each award is compared with its budget. A subcontract awarded below budget creates a buyout saving; one awarded above creates a loss. Tamarack's policy was to move buyout savings into the contingency line rather than treat them as extra profit, until the job was safely past its riskiest stages. A capstone plan should describe this policy, because it shows how the contractor protects itself against later surprises.

The cost-loaded schedule

Cost loading assigns each activity in the schedule its share of the contract value. Spread over the activity's duration, those values give the planned value earned each week and each month. Added up over time, they form an S-curve: slow at the start during mobilization and earthwork, steep in the middle when steel, envelope and rough-in run together, and slow again at the end during finishes and commissioning.

Tomas used the base 60-week network rather than the 56-week plan. A slower schedule stretches the period during which Tamarack carries costs before it is paid, so it is the more cautious case for cash flow.

Progress billing and money held back

Under the contract, Tamarack would apply for payment once a month for the value of work completed, using a schedule of values that breaks the contract price into lines the Consultant can check on site. The owner pays the certified amount less a holdback of 10 %, as provincial lien legislation requires in most provinces. The holdback protects subtrades and suppliers who might register liens and is released after substantial performance, once the lien period has passed with no claims. Tamarack assumed that each payment would arrive about a month after the application, as the contract's payment terms allowed.

On the other side, Tamarack pays its own staff weekly, its suppliers within thirty days or so and its subtrades soon after it is paid, also holding back a share from each subtrade. Even with that, the general contractor's own costs and the timing of payments mean that cash usually goes out before it comes in. Tomas built a simple model to estimate how far out of pocket Tamarack would be.

# Monthly earned value, billing and contractor cash position (thousands of dollars)
import math
acts = {  # id: (start week, duration weeks, budget in $000)
    'A': (0, 3, 420), 'B': (3, 5, 980), 'C': (8, 6, 1250), 'D': (8, 5, 760),
    'E': (14, 6, 1350), 'F': (20, 5, 1010), 'G': (20, 9, 1500), 'H': (25, 10, 1900),
    'I': (29, 3, 520), 'J': (35, 8, 1400), 'K': (43, 7, 1030), 'L': (29, 6, 760),
    'M': (50, 4, 482), 'N': (54, 4, 260), 'O': (58, 2, 190),
}
weeks = 60
weekly = [0.0] * weeks
for s, d, b in acts.values():
    for w in range(s, s + d):
        weekly[w] += b / d
months = math.ceil(weeks / 4.3)
earned = [sum(weekly[w] for w in range(weeks) if int(w / 4.3) == m) for m in range(months)]
HOLD = 0.10                  # holdback withheld from each payment
cost_ratio = 0.95              # contractor's costs are about 95% of earned value
cum_ev = cash = low = 0
print('Month  Earned  Cum.earned  Paid in  Cash position')
for m in range(months + 1):
    ev = earned[m] if m < months else 0
    paid = earned[m - 1] * (1 - HOLD) if m >= 1 else 0   # paid one month after billing
    cum_ev += ev
    cash += paid - ev * cost_ratio
    low = min(low, cash)
    print(f'{m + 1:5d} {ev:7.0f} {cum_ev:11.0f} {paid:8.0f} {cash:14.0f}')
total = sum(earned)
print(f'Total {total:,.0f}; holdback held at the end {total * HOLD:,.0f}; lowest cash position {low:,.0f}')

Month  Earned  Cum.earned  Paid in  Cash position
    1     812         812        0           -771
    2     948        1760      731           -942
    3    1441        3202      854          -1457
    4    1108        4310     1297          -1213
    5    1187        5497      998          -1343
    6    1463        6960     1069          -1664
    7    2050        9010     1316          -2295
    8    1440       10450     1845          -1818
    9     700       11150     1296          -1187
   10     700       11850      630          -1222
   11     736       12586      630          -1291
   12     535       13121      662          -1138
   13     371       13492      482          -1008
   14     320       13812      334           -979
   15       0       13812      288           -691
Total 13,812; holdback held at the end 1,381; lowest cash position -2,295

The model shows the lowest cash position of about −$2.3 million in month 7, when steel, envelope and rough-in are all being billed and paid for at once, and about −$0.7 million at the end of month 15, before the holdback of about $1.38 million is released. Once the holdback is paid, Tamarack's cash position turns positive by roughly the fee it planned to earn. The model is crude: it treats Tamarack's costs as 95 % of earned value, ignores the holdback Tamarack itself retains from subtrades and assumes payments arrive on time. Even so, it tells management how much working capital the project will tie up, and it shows why prompt payment and early release of holdback matter so much to contractors.

Ruth used the result to make two decisions. She asked the finance team to confirm that Tamarack's bank line could carry the project alongside its other work, and she asked Farah to make front-end loading off-limits. Front-end loading means putting too much value into early schedule of values lines so that the contractor is overpaid at the start. It improves cash flow, but owners and consultants watch for it, and it destroys trust when it is found. A balanced schedule of values, checked against the cost-loaded schedule, was the honest alternative.

Earned value: measuring cost and schedule together

Once construction starts, the cost-loaded schedule becomes the baseline for measuring performance. Earned value management compares three numbers at each reporting date: planned value (the budgeted cost of the work scheduled to date), earned value (the budgeted cost of the work actually done) and actual cost. From them come two simple indices.

An SPI below 1 means the project is behind its planned progress. A CPI below 1 means the work done has cost more than budgeted. The estimate at completion, EAC, projects the final cost if the current cost performance continues. For a contractor, it is most useful to measure these against the cost budget (the bid less the fee) rather than the contract price, because the question that matters is how much of the fee is being consumed.

A forecast like that is not a verdict. It is a warning that triggers action: finding which cost codes are overrunning, whether the cause is a change the owner should pay for, a productivity problem or a pricing error, and what can be done about it. CON-310 covered the detailed methods. The capstone plan should state which indices will be reported, how often, at what thresholds the project team must explain a variance, and who acts.

The cost control plan

A cost control plan describes the routine that keeps the budget honest. It is not long, but it should be specific. Tamarack's plan for Pembrook had six parts.

• Coding: every cost charged to a cost code within two days of the invoice or timesheet.
• Commitments: subcontracts and purchase orders recorded as committed cost when signed, not when invoiced, so the forecast sees them early.
• Monthly cost report: budget, committed cost, cost to date, forecast to complete and forecast at completion for every code, reviewed by the project manager and Ruth.
• Changes: every change tracked from first notice to approved change order, with its cost kept separate from base contract costs.
• Contingency: a running record of contingency used, with a reason for each draw.
• Thresholds: any cost code forecast to overrun by more than $25,000 or 5 % is explained in writing with a recovery plan.

The monthly report also forecast cash. Each month, the project team updated the cash flow model with actual billings and payments, so that Tamarack's finance team could see the project's real demand on working capital rather than the planned one.

Changes and their effect on cash

Changes affect cash as well as cost. When the owner directs extra work, the contractor usually pays for it before it is priced and approved, and sometimes long before it is paid. On Pembrook, Farah planned to price every change within ten working days of a request and to bill approved changes in the next monthly application. For work done under a change directive before the price was agreed, Tamarack would bill the undisputed part and record the rest as a pending claim. CON-420 covered the rules for changes and claims; the capstone plan should show how changes flow into the cost report and the cash forecast.

Your capstone deliverable

• A control budget by cost code, with the fee and contingency shown separately.
• A cost-loaded schedule and S-curve of planned value.
• A cash flow model showing billings, payments, money held back and the contractor's cash position.
• An earned value reporting plan with the indices you will report and the thresholds for action.
• A cost control plan covering coding, commitments, the monthly report, changes and contingency.