Cost management is the work of estimating what a project will cost, turning the estimates into a baseline spread across time and then comparing what has been spent against that baseline and acting on the difference. Earlier editions of the PMBOK Guide called it cost management and treated it as a knowledge area. PMBOK 8 folds it into the Finance performance domain, alongside funding, benefits and the financial case, and the examination tests it through reserve questions and through the reading of a cost position.
The three activities, estimating, budgeting and control, get collapsed into one in ordinary conversation, and keeping them apart is most of the subject. The sections below take them in order, then cover the two reserves, direct and indirect cost, what happens to a budget with no reserve and the reconciliation of spend against the money actually available.
Estimating the cost
Four techniques produce a cost estimate, and they are the same four used for duration. Analogous estimating scales a completed project of similar shape. It is quick, available before anyone understands the work, and it inherits every peculiarity of the project it borrowed from. Parametric estimating multiplies a measured rate by a quantity, and it is exactly as good as the rate and the quantity are. Bottom up estimating costs each package of work and adds them up, which is the most accurate method available and impossible until the work has been decomposed.
Three point estimating asks for an optimistic, a most likely and a pessimistic figure and combines them. It is the only one of the four that returns a shape, and that matters because the figure reaching the sponsor is one value pulled out of a distribution. Handing over the mean without the spread throws away what the sponsor needed in order to decide how much cover to hold.
The cost baseline and the funding requirement
Aggregating the estimates gives the cost of the work. Adding contingency reserve gives the cost baseline, which is the authorised, time phased spend the project is measured against. Adding management reserve on top gives the total project budget, and the funding requirement is that budget laid out against the dates on which the money actually has to be available.
Estimates aggregate into the cost of the work, and adding contingency reserve gives the cost baseline that every variance is calculated from. Management reserve sits above that line and inside the total budget, which is why releasing any of it moves the baseline.
The distinction decides what counts as a variance. The baseline is what variance is calculated from, so everything inside it is spend the project promised to make, and money released from above the line arrives as a change to the promise.
Contingency reserve and management reserve
The two reserves in the diagram cover different things and belong to different people. Contingency reserve covers identified risks. It lives inside the cost baseline because those risks were analysed and priced when the baseline was built, and the project manager draws on it without asking permission.
Management reserve covers what was never identified. It sits outside the cost baseline and inside the total budget, it belongs to the sponsor or the governing body, and releasing any of it is a change that moves the baseline. That is the practical test for telling the two apart.
Direct and indirect cost
The estimates inside the baseline are of two kinds. A direct cost is incurred for this project and would not exist without it, such as the people assigned to it and the licences bought for it. An indirect cost is shared across several pieces of work and allocated to each, such as premises, shared services and management overhead.
Indirect cost matters to a project manager because it is allocated by a formula nobody on the project chose, so it moves for reasons unconnected to how the project is performing. An overrun caused by a change in the overhead recovery rate is a real overrun that arrived from the finance system, and a project manager who cannot separate it from delivery spends the review defending work that was never the problem.
The budget with no reserve
The reserves are the part of the budget most often argued away, and a budget with no reserve is a forecast presented as a commitment. Every estimate carries uncertainty, so a total built out of estimates carries it too. Stripping the reserve out leaves the uncertainty exactly where it was and removes the funded response to it, and the first risk that materialises then gets paid for out of scope, out of quality or out of somebody else's project.
This is why a reserve deleted in a budget negotiation reappears later as a change request, an unpaid overtime bill or a quietly lowered acceptance standard. The honest version of that conversation states the point estimate, states the confidence attached to it and names the amount being held against the risks already on the register. A sponsor is entitled to refuse the reserve. No sponsor is entitled to refuse the reserve and keep the confidence level that came with it.
Funding limit reconciliation
Even a budget with its reserves intact is rarely available all at once. Money is released annually, quarterly or at agreed gates, and each period has a ceiling. Funding limit reconciliation compares planned spend against those ceilings period by period and reschedules work when the plan wants more money than the period allows.
It is a scheduling constraint that arrives from finance. Where the plan spends 400,000 in a quarter that has 300,000 available, something moves, and moving it changes the schedule and usually the cost. Finding that out while the baseline is still being set makes it an ordinary planning decision, and finding it out when an invoice cannot be paid makes it an incident.