Initial Benchmarking of Highways England’s Regional Maintenance Spending

Initial Benchmarking of Highways England’s Regional Maintenance Spending

Nominated for The Neil Mansfield Award


Adam Spencer-Bickle, Office of Rail and Road


This paper sets out the ORR’s initial work to benchmark Highways England’s regional maintenance spending, through a mixture of Corrected Ordinary Least Squares and Stochastic Frontier Analysis models.


The Office of Rail and Road (ORR) took on the role of Highways Monitor in April 2015. There are two key aspects to the role: monitoring Highways England’s delivery of the requirements in the Road Investment Strategy (RIS), and providing advice on the degree of challenge and deliverability in future strategies.
This advice is primarily in the form of an Efficiency Review of Highways England’s plans, which must consider the overall level of challenge and deliverability within the available funds, with a specific focus on the levels of efficiency that Highways England proposes to achieve. The first RIS covers 2015-2020 and development of the second, RIS2, is already underway.
As with many other regulators, benchmarking will form an important part of our Efficiency Review. This could take many forms, from top-down to bottom-up, looking at efficiency and levels of performance. When looking at efficiency, we are planning to develop top-down econometric cost/efficiency models similar to those used in other sectors, including by ORR for Periodic Reviews of Network Rail.
In some sectors, like water, where there are many similar companies (or regional monopolies) regulated by the same regulator, it is relatively straightforward to collect the data needed to populate these models. However, for national monopolies, like Highways England and Network Rail, choosing comparators and collecting data are more challenging.
For rail, ORR has used the Lasting Infrastructure Cost Benchmarking database to benchmark Network Rail’s costs against other national rail infrastructure operators in previous Periodic Reviews. However, international benchmarking like this creates its own challenges – differences in standards, input prices and exchange rate fluctuations are just some of the factors that can affect cost differences across countries. Therefore, and in line with the move to more route-level regulation in PR18, this will be complemented with more route-level and internal benchmarking of Network Rail’s costs.
On the road side, an international database of consistently defined costs and explanatory variables for ‘strategic road networks’ is not available. And setting up a benchmarking network or ‘club’ would likely be a significant, long-term undertaking. So we are planning to focus our top-down cost benchmarking of Highways England for RIS2 on internal benchmarking of its regions.
This paper provides an overview of the approach taken to modelling rail costs in previous Periodic Reviews, to give context and introduce the set of Corrected Ordinary Least Squares (COLS) and Stochastic Frontier Analysis (SFA) models used in the highways analysis. It then applies these COLS and SFA models to Highways England regional maintenance costs, and compares the efficiency results with simpler unit cost analysis. Given the ‘in progress’ nature of the work and emerging findings, and that benchmarking is only one part of the RIS2 Efficiency Review, these initial results do not represent ORR’s view on Highways England’s level of efficiency, or further efficiency improvements that could be made in RIS2. Finally, the paper discusses how the data and analysis could be developed further, including by expanding the analysis to include regional data from National Road Authorities in other countries.


Association for European Transport