Guides · Costs and funding
Fixed recoverable costs: three ways out, three different tests
Applications to get out of fixed costs are routinely argued as though there were one question. There are three, they sit at different points in the life of the case, and the thresholds are not close to each other. Picking the wrong one is usually why the application fails.
The three routes are not alternatives
Exceptional circumstances, CPR 45.9(1). The court may consider a claim for costs greater than the fixed recoverable costs where there are exceptional circumstances making it appropriate. The rule offers no guidance on what qualifies, which is deliberate, and the bar sits accordingly high. That a case was hard fought, or that the work genuinely done bears no sensible relationship to the figure in the table, is not on its own exceptional. Something has to have taken the case outside the run of cases the band was written for.
Reallocation to a different track, CPR 26.18(2). Once a claim is on the intermediate track and directions have been given, the court may reallocate only where there are exceptional reasons to justify doing so. Before directions, the ordinary allocation criteria apply and the argument is a normal one.
Reassignment to a different complexity band, CPR 26.18(3). This requires a change in circumstances since a direction was made, and a conclusion that the change justifies reassignment. That is a materially lower bar than either of the others, and it is the route most often left on the table.
The thresholds fall as you go down that list, but so does the prize. Reassignment moves you within the same costs table. Reallocation moves you into a different costs regime altogether, and CPR 46.13 makes it do so retrospectively: the costs which may be allowed are those applicable to the track the claim is reallocated to, as if the claim had been allocated to that track at the outset.
The retrospective effect of reallocation is the part that gets missed. It does not draw a line at the date of the order and apply the new regime going forward. It reaches back over work already done and already billed, which is what makes it worth more than a band reassignment, and worth arguing for even late.
The 20% rule can make a successful application worse than no application
CPR 45.9(2) is where escaping fixed costs turns into a trap. Succeed on exceptional circumstances, go to assessment, and if the assessed costs come in at less than 20% above the fixed recoverable costs, the court will allow the lower of the assessed costs and the fixed recoverable costs, and may order the applicant to pay the costs of the assessment.
So the question is not whether you can beat the fixed figure. It is whether you can beat it by a margin, having also absorbed the cost of an assessment you may then be ordered to pay for. An application that succeeds on the merits and lands at 15% above the table leaves the receiving party holding fixed costs and an adverse costs order on the assessment.
Contrast the structure of the vulnerability provision at CPR 45.10, which puts the same 20% figure in a completely different place. There, the additional work caused by vulnerability must, by itself, produce a claim at least 20% above fixed recoverable costs before the route opens at all. One rule gates entry on the margin, the other gates the reward on it. Reading across from one to the other produces the wrong answer both ways.
Argue the band before you need to argue your way out of it
The cheapest of these arguments is the one made at allocation, and it is consistently underprepared because it happens before anyone has felt the cost of the band they were put in. Complexity is far easier to establish prospectively, on the shape of the pleaded case, than retrospectively against a court that can see exactly how the case turned out.
The change in circumstances route has a quiet timing problem too. The change must post-date the direction, so the longer a party waits the more material there is to point at, and the more of the costs being complained about have already been incurred under the band that is said to be wrong. Waiting builds the argument and shrinks its value at the same rate.