Capex Decisions in a High-Cost-of-Capital World — When IRR Is Not Enough

The Projects That Made Sense at 4% Don't All Make Sense at 5.5%

Between 2020 and 2022, the cost of capital for most businesses was at historic lows. Projects that generated IRRs of 6-8% were approved because they cleared the hurdle rate comfortably. Many of those projects are now being executed with debt costs 150-200 basis points higher than when they were approved. Some of them no longer clear the hurdle rate at current cost of capital.

This isn't hypothetical. In our work with a tire manufacturer on a $50M capex portfolio, we found three projects in execution that had original business cases built at a 4% WACC. At the current 5.5% cost of capital, two of those projects had IRRs that no longer exceeded the hurdle rate. The organisation was investing capital in projects that were actively destroying value relative to their cost — and no one had done the reassessment.

The $50M Misallocation: What We Found

The capex portfolio had been approved over a 3-year period with different financing assumptions, different revenue assumptions, and different risk profiles. The problem was that no one was looking at the portfolio as a whole, reassessing it annually, or applying a consistent methodology across all projects.

We built a portfolio review that assessed every active capital project against three cost of capital scenarios: 4% (the original approval environment), 5.5% (current), and 7% (stress case if rates rise further). Projects were colour-coded: green if they exceeded the hurdle rate in all three scenarios, amber if they exceeded in base but failed in stress, red if they failed at current rates.

The distribution was striking. Of the six major active projects, two were green across all scenarios, two were amber, and two were red at current cost of capital. The two red projects represented $18M of the $50M portfolio. The recommendation was to pause them pending renegotiation of scope or financing terms. Continuing to invest in projects with negative risk-adjusted NPV at current rates was not capital discipline — it was inertia.

WBS as a Control Layer

One of the structural problems in the capex review was that project costs were not tracked at the Work Breakdown Structure level. Capex was approved at the project level, and actual spend was tracked at the project level. This meant that cost overruns in one work package were invisible until they grew large enough to affect the total project line.

Implementing WBS-level tracking — requiring each capital project to define cost categories (civil, mechanical, electrical, commissioning, contingency) and track actuals against each — created an early warning system. A civil works package running 30% over budget in month two was a signal that total project cost estimates might need revision, long before the overrun showed up in the headline number.