The Scottish Government

« Previous | Contents | Next »

Listen

ANALYSIS OF HISTORICAL CONSTRUCTION COST MOVEMENTS IN SCOTTISH SOCIAL HOUSING FINAL REPORT

3.0 MAINTENANCE COSTS

3.1 Introduction

This section of the report examines historic trends associated with planned and reactive maintenance works.

In 2001/2 approximately 700 per dwelling was spent on maintenance by Scottish RSLs, about 60% on planned, cyclical and major repairs and 40% on reactive maintenance. (Source: Communities Scotland)

Maintenance cost indices measure the cost of the inputs to maintenance contractors. Derived using similar methods and sources of base cost information to the Building Input indices discussed under Capital Costs, these series again do not take account of overheads and profit variation, wage levels actually paid to operatives or changes in achieved productivity.

Two principal sources of data have been used in this section:

• Department of Trade and Industry (DTI) - Resource cost indices of maintenance
• Building Maintenance Information (BMI) - Maintenance cost indices

3.2 Department of Trade and Industry Indices

3.2.1 Housing and non-housing

The DTI publishes a series of UK Resource Cost Indexes of Maintenance in ' Quarterly Building Price & Cost Indices'. As before, DLE have not been able to obtain access to details of the models or weightings used.

For maintenance work, the DTI produces two series of indices, for Non-housing and for House Building:

Resource Cost Index of Maintenance for Building: Non-Housing -NOMACOS
Resource Cost Index of Maintenance for House Building - HOMACOS

Chart 3.1 plots these two index series, deflated to reflect real cost increases, between 1985 and 2002. Over the whole period, the two trends have moved closely in tandem, HOMACOS increasing by an average annual percentage increase of 0.3% and NOMACOS by 0.4%.

Chart 3.1 Resource Cost Indices for Building Maintenance

In the early part of the period, 1985-1992, real maintenance costs fell relative to general inflationary trends but have been on an increasing trend since. As noted under Capital Costs, this is largely due to building industry wage awards exceeding wage growth in other sectors, particularly since 1999.

With no significant divergence at any time over the 17 year period, evidence suggests that the cost of maintenance for housing work and for non-housing work tends to move in unison.

3.2.2 Sub-index examinations

Examination was undertaken of differential cost movement in sub-indices relating to:

• Building maintenance work
• Mechanical maintenance work
• Electrical maintenance work

Analysis shows similar patterns to those identified under Capital Costs, generally consistent increases over time, the detailed cost trends largely dictated by the timing of wage awards in the three industries.

Further study was undertaken of the split between labour and plant and materials costs. Over the period, real costs of maintenance materials have been in general decline (except for 1992 to 1995) and have fallen by 15%. In contrast, labour costs have experienced only minor variations from general inflation until 1997, when real labour costs began to increase sharply (at an average annual rate of 4.3% between 1997 and 2002).

Chart 3.2 Labour & Plant and Materials Components of HOMACOS

This pattern of costs is similar to that observed in Chart 2.3, which examined the real cost trends of labour and plant and materials for new build housing.

Examination of the real cost movements for labour and plant and materials for Maintenance for Building: Non-Housing (NOMACOS) revealed a similar pattern of change.

3.3 Building Maintenance Information indices

3.3.1 Introduction

Building Maintenance Information is a service of the RICS Building Cost Information Service. BMI compile and publish a series of Building Maintenance Cost indices in ' BMI Quarterly Cost Briefing'.

The principal index series are input cost indices, reflecting the cost to the contractor of carrying out maintenance work and are based on the costs of labour, materials and plant. The series consist of four groups of indices relating to different employment sectors:

• Health Service
• Local Authorities
• Private Sector
• 'All-in'

The 'All-in' index is based on a combination of the three different employment sector indices. For this research, the Health Service series of indices have not been considered but the All-In, Private Sector and Local Authority series of indices have all been examined.

For each series of indices, BMI produce three sub-indices and a general index relating to various types of work:

• Redecorations
• Fabric maintenance
• Services maintenance
• General maintenance

The General Maintenance indices are a combination of the other three sub-indices: redecorations, fabric and services maintenance.

It should be noted that BMI do not produce a maintenance cost index specific to housing or building type.

BMI have published details of the composition of their indices in ' Occupancy Cost Indices: Description, Methodology and Background'.

The BMI indices have been examined to confirm the consistency of the trends observed from DTI indices and to identify the principal factors driving cost movements. The breakdown of the two index source data into different component parts provides an opportunity to identify differentials in cost trends from the broadest possible perspective. Used in conjunction with the Social Housing Refurbishment index prepared by DLE, verification of the appropriateness of the application of these indices to Scotland can be established.

3.3.2 Employer differentials

Chart 3.3 Building Maintenance Information - Cost Differentials by Employer

Chart 3.3 illustrates real cost movements recorded by BMI's General Maintenance Cost Indices in respect of Private Sector Maintenance, Local Authority Maintenance and their All-in Index. Between 1985 and 1997 the index measuring real Local Authority Maintenance Costs rose by 10.4% compared to a real rise of 3.6% for Private Sector Maintenance.

The pattern changes markedly after 1997, from which point the private sector index shows a much steeper rate of increase, related to the size of wage awards received by private sector operatives over the last five years. Building workers tied to local authority pay scales have not been in receipt of similar rewards and the index actually fell, compared to general inflation, between 1996 and 1999 and has done little but match inflation since.

This finding suggests that wage pressures affecting work on schemes transferred to the RSL sector will be greater than those that have been experienced in the DLO sector, where real maintenance costs have risen by an average 0.7% per annum less than the private sector over the study period.

3.3.3 Types of maintenance work

BMI indices are also available for different areas of maintenance, tracking costs of redecorating, fabric and services maintenance. Chart 3.4 plots the real cost movement of these components, from BMI's All-in Maintenance Cost Indices series, from 1985.

Chart 3.4 Building Maintenance Information - Cost Differentials by Type of Work

Fabric maintenance has shown the lowest real cost increase throughout the whole period, rising 9.9% over the 17 years (or an average 0.6% per annum). Redecorations has shown the greatest real cost increase over the whole period, rising 18.9% (or 1.0% per annum) but rising to 2.0% per annum for the later period since 1997. This reflects the fact that redecorations is the most labour intensive of the three basic types of maintenance, having a base labour weighting of 89% compared to 65% for both fabric and services maintenance. Labour cost has been shown elsewhere to be the prime cost driver in the period post-1997.

Similar studies were undertaken of the Private Sector and Local Authorities series of maintenance indices with similar patterns of results.

This pattern shows that long term redecoration liabilities, works that need to be undertaken on a reasonably regular basis, will be under marginally greater pressure in terms of input costs than fabric and services, which may be renewed at an early stage of a transfer programme.

3.4 Maintenance Cost Model specific to Scottish Social Housing Maintenance

As described in section 2, a model has been developed specifically to relate to the work anticipated to be undertaken under the heading of component replacement over the next 30 years. The exercise shows that there was little difference in cost movement over the last 11 years between the specific model and generic data published by DTI and BCIS.

A similar exercise could be carried out to develop a maintenance cost model specific to Scottish Social Housing, based on the actual maintenance work carried out on housing stock recently or on anticipated maintenance workloads. However, due to the close correlation between the specific and generic models under Capital Costs and the considerable effort required to construct one-off indices, it is considered that no new trends are likely to be identified with a specifically constructed maintenance cost model and consequently, this exercise has not been undertaken.

3.5 Summary

Table 3.1 summarizes the maintenance cost trends identified in this research.

It is clear from the table that the real cost increase indicated by DTI indices (HOMACOS and NOMACOS) over the period examined is lower than that shown by BMI indices. Without access to the DTI cost models, it is not possible to be certain of the reasons for this divergence but it would seem likely that the BMI models contain a higher weighting of labour.

Table 3.1 Summary of Maintenance Cost Indices and Real Percentage Changes

3.6Capital and Maintenance Costs Compared

Chart 3.5 compares the index series identified as being the most suitable for comparing capital and maintenance costs. The Resource Cost Index of House Building (HOCOS) is plotted against the Resource Cost Index of Maintenance for House Building (HOMACOS), the latter being the only maintenance cost index specifically related to housing, to examine the extent of variation in costs between new build/refurbishment and maintenance work. All the indices in this chart are expressed in real terms.

The real cost increases plotted in the graph show that the resource costs for new build housing and maintenance of housing (based on DTI indices) have never differed by more than a percentage point or two. From 1985 until 1998, new build costs had shown a slightly greater rate of increase than maintenance costs. Since 1994 the rate of real maintenance cost increases has been slightly ahead of new build increases: this is due to the higher labour component, reflecting the general increases in construction labour costs described earlier in the report (and detailed in Chapter 9).

Chart 3.5 Resource Cost Indices for House Building & Maintenance for House Building

Chart 3.6 compares the movement of the cost model evolved for refurbishment work on Scottish Social Housing with the Resource Cost Index of Maintenance for House Building (HOMACOS) from 1991.

The chart and table show that, over the study period from 1991, real cost increases for Scottish social housing refurbishment and U.K. housing maintenance costs have been almost identical. Over the first half of the period, to 1997, refurbishment costs increased at a slightly lower rate (average 0.4% per annum) than maintenance costs (average 0.8% per annum). This pattern has reversed during the latter half of the period, the former increasing at an average annual rate of 2.2% per annum compared to 1.8% for the latter. Both of these index series indicate that the rate of growth has accelerated since 1997, driven by recent labour cost increases.

Chart 3.6 Cost Indices for Refurbishment and Maintenance of House Building

« Previous | Contents | Next »