Analysis of previous increases in HGV sizes and efficiency of existing sized lorries- MTRU
Introduction and context
This report updates the basic trend analysis undertaken in 2007/2008 for the MTRU report Heavier lorries and their impacts on the economy and the environment for Freight on Rail. It adds some new information to show the impact of the recession.
Most of the available data for the original report was from 2006. Data from 2007 and 2008 is now available from Transport Statistics Great Britain and from 2009 from the Continuing Survey of Road Goods Transport (CSRGT) as reported in Road Freight Statistics 2009.
Impact of the recession
The last two years are clearly reflecting changes brought about by the recession, and this can be seen in a new chart, showing changes in vehicle kilometres for all HGVs over the last decade. This includes goods vehicles down to 3.5 tonnes, but illustrates the serious decline. It uses data from DfT counters. The CSRGT uses a different method for estimating goods vehicle activity which is also shown on the chart, but this too shows a severe decline.
Figure 1A (not in 2008 report)
In addition, the recession has hit the public haulage sector particularly hard. Companies which operate their own heavy vehicles (own account) appear to have maintained activity. This may well be due to their transferring work from outside hauliers to compensate for an overall drop in demand. Public hauliers are familiar with this compensating mechanism from previous recessions. The following chart is reproduced from Road Freight Statistics 2009.
Figure 1B (not in 2008 report)
The next section updates the figures in the 2008 report, although the general decline reflected in the above charts means they have to be interpreted carefully. Having said that, the trends identified in 2007/8 have not changed as much as might be expected and the overall conclusions are still valid.
The particular issue of why the heaviest lorries often run empty (by volume or weight) or part loaded by weight, is explored further in a new Annex. This draws on special tabulations requested from the CSRGT.
2. Updating the 2008 report
Falling average payloads by weight
The first measure considered by the 2008 report was how much freight (in terms of tonnes) the average articulated HGV (artic) carries. This is simply calculated from CSRGT by dividing the survey tonne kilometres by the survey vehicle kilometres. If the increased limits had allowed more goods to be carried by the same number of vehicle kilometres, this should show up as an increase in the average payload in the following years. The exercise was undertaken for the type of vehicles which were subject to the increases: artics with a gross permitted weight of 33 tonnes or more.
In fact the average payload has fallen considerably since 1983, with very little evidence of even short term increases. Since 1995 there has been relative stability, although weight limits have increased twice, allowing an extra 6.5 to 7 tonnes payload on 6 axles. This is shown in Figure 2 below which has now been updated.
Figure 2 from 2008 report, updated with 2006 to 2009 data
Figure 2 shows that, following a distinct fall in average load by weight from the mid 1980s to the mid 1990s, this has stayed fairly stable. This was despite significant increases in maximum permitted weight in 1999 (38 to 40/41 tonnes) and 2001 (up to 44 tonnes). For the weight increases to deliver efficiency improvements, the average should have risen. The recession does not appear to have had a major impact.
This analysis is based on weight increases, but there is an important alternative measure for loading – volume. This relates to low density goods which may fill a vehicle before its maximum weight is reached. This is often referred to as volume constraint or “cubing out”. This effect has been discussed at least as far back as the 1980s. It is very variable, for example some consumer goods may be low density, such as paper products, although some are dense, for example domestic appliances. Construction, chemical and agricultural materials tend to be dense, while specialist commodities like cigarettes are lighter for a given volume.
Direct and robust evidence on the overall extent of this problem and how it has changed over time is not available, but there are several indirect ways of trying to assess whether this is the true explanation for the fall in the weight of average loads.
Can “cubing out” explain falling payloads by weight?
The first problem with the cubing out explanation is that there are vehicle options which would allow a greater volume for the same gross weight. One example is draw bar trailers, which can carry up to 44 tonnes. These can be up to 18.75 metres long and are articulated in the middle of the vehicle rather than at the front just behind the driver’s cab. Demand for such vehicles has continued to be very low throughout the period studied here. They are not suitable for container traffic, and split any load into two sections, unlike standard tractor/trailer combinations. However they can utilise swap bodies offering similar flexibility to the trailers used by traditional artics.
More straightforwardly, maximum length, lighter artics on 4 axles could be used to give a much higher volume to weight ratio. They would be cheaper to operate and produce less CO2. However, this category of vehicle has in fact declined in number very significantly, from 42 thousand in 1990 to 12 thousand in 2008 1.
There is another way in which cubing out can addressed – the use of “double decking”. When pallets are used, goods loaded on them may not reach the full height of the load area. Thus they are constrained by the floor area rather than volume. Creating a second floor in the trailer allows better utilisation. However, if this were widespread, the average payload should have risen rather than fallen. This indicates that cubing out may not be as critical as some claim.
A further reason for believing that volume constraint is not the only answer to the falling payloads is the increase in dimensions in 1990. This was very significant at about 8 to 10%. If volume was a problem this would have had a major impact and should have allowed average payloads to rise, at least in the short term. This simply did not happen. The most significant rise in payload (still small) occurred after the 2% increase in width in 1996 and lasted one year.
Matching vehicles to the loads
One possible explanation offered in the original report was that public hauliers in particular would purchase the largest vehicle they were ever likely to need, and use it for their existing business which had previously been undertaken in HGVs with a lower maximum permitted weight. This would be explained if a 38 tonner was replaced with a 40 or 44 tonner, and the same consignments were transported. The rapid uptake of the maximum sized vehicles is shown in Figure 3, although it should be noted that in many cases the 40 tonners were capable of carrying 44 tonnes and thus the move was simply a matter of replating the vehicle.
Figure 3 from 2008 report, revised to show all artics and updated from 2006 to 2008
Did size and weight increases reduce traffic?
One further question posed in the original report was whether the increases in size and weight had achieved the reductions in HGV traffic which were claimed. This is difficult to assess because traffic normally varies from year to year, but there is no discernible impact, for example after 1983 when the maximum weight was raised from 32.5 to 38 tonnes. In reality, the level of HGV activity during these years is probably more strongly influenced by the economic cycle.
Figure 4 from 2008 report updated from 2006 to 2008
Further light is shed on this by looking at the figure for actual tonne kilometres compared to capacity tonne kilometres calculated for the CSRGT. This shown in Figure 5 below.
Figure 5 from 2008 report, updated
A new chart, showing the proportion of time spent completely empty, has also been produced to illustrate how this has shown very little variation over time. It appears to have worsened at the start of the recession before falling back slightly in 2009.
Finally, the impact of these charts was brought together with a table listing the main changes to maximum permitted weights and dimensions. The table is reproduced below and the summary chart (revised Figure 6) follows.
The conclusion from Figure 6 must be that there is no discernible link between changes in size and weight limits and either:
The impact of the recession has led to a major fall in the public haulage sector but far less so in own account operators. This is understandable where the highly flexible public haulage sector is used to cope with peak demands which own account operators cannot meet. Overall there has been a significant fall in vehicle kilometres by HGVs. The latest year’s figures show a slight improvement in average vehicle load, but no improvement over the long term average for empty running.
The issue of whether average weight has decreased due to goods becoming less dense, and thus reaching a volume limit, was discussed in the original report. In broad terms, there has been some effect for certain commodities, but the data is not available to correlate this with other changes. It is also the case that high volume vehicle options have not proved popular.
On weight related measures, the most recent figures show little reason to alter the first three key findings of the original report, which were:
“After examining the most reliable sources of national statistics, the conclusions are:
This note reflects further upon the low average loads by weight in the heaviest HGVs (which are articulated), compared to capacity which are described earlier in the update report. This has been evident for several decades, despite various changes in maximum permitted weights and sizes. It focusses on the articulated vehicles which dominate the longer distance freight market and uses new tables supplied by the Continuing Survey of Road Goods Transport (CSRGT).
Weight, volume and floor area
National statistics for UK registered vehicles have tracked the average payload by weight, and the number of kilometres that articulated HGVs run completely empty, for a considerable time through the CSRGT. There are no comparable figures for payload by volume - in other words when the total vehicle space is used up but the load is below the maximum permitted. This effect is often said to be caused by lower density goods is known as “cubing out”.
However, this is not the complete picture, nor even an accurate description, because there are various standardised ways of packing goods, for example pallets or cages, which can come up against the limits of the vehicle’s floor area available for loading, rather than volume.
One interesting way of dealing with this has been using a double deck arrangement within the HGV’s trailer. although this is not very widespread at present. It has its own complications, for example accessing the second floor level and dealing with pallets of uneven heights.
Volume is an area where there is ongoing work to establish an overall picture.
1.Meeting the single load requirements
The first is that the industry has standardised its longer distance fleet on the largest articulated vehicles which are permitted. This is clear from the vehicle registration data, with most articulated vehicles in the maximum weight category. The business model is to buy vehicles which will be able to compete for any weight of load.
In this case there is very low flexibility to match vehicles to individual loads and the expectation would be many single drop loads well below the maximum permitted. The new figures show some evidence to support this.
Figure A1: Distance travelled by weight being carried
Within this, there is an even spread of very light loads, as shown below.
Figure A2: Distance travelled with lowest loads (excluding empty running)
2. Grouping loads together
In the situation where demand is to transport loads lighter than the maximum permitted, and only the largest vehicles are available,one way to regain efficiency is to group together (consolidate) smaller loads with other loads going to destinations anywhere along the route. Return loads can also be sought, reducing overall vehicle kilometres and resource costs. There are systems and software which handle this already available, sometimes using a clearing house to put operators in touch with each other.
There are barriers to this however, such as a need for speed of delivery, high security, or compatibility of loads. In the case of public hauliers, the motive to overcome these barriers will be weaker in a market with many providers who are competing strongly, as is the case in the UK. Competition can be high, but efficiency low. This can be of course be avoided at least in part by including the environmental and congestion costs into the market, but this is only partially done at present and at a low level.
Of course, there are a different set of barriers in the own account sector, where heavy vehicles are owned by a specific company and dedicated to their operations. These find it easier to consolidate for the outward journey, but tend not to carry much (if anything) on the return journey to their company’s depot (back loading). There has been some progress on this, for example supermarkets taking consolidated loads to stores, then picking up goods from suppliers to be taken to their distribution depots.
A further explanation may be that third party consolidation services have simply not been fully developed in the UK. This will probably also vary between individual firms and between own account and public operators. However, the potential for such services to reduce vehicle kilometres and thus emissions, congestion and environmental damage, is considerable. German experience of reducing empty running over the last decade is a case study of what can be achieved - reducing it from levels comparable to the UK in the 1990s to over 20% less today.
If loads are consolidated the data would show a larger number of drops per journey plus lower empty running. The pattern of loads would be a high amount at maximum load by weight (made up of different deliveries), with a steady tapering down towards the lighter end as the loads were delivered.
In fact, the low level of 2 to 5 drops per journey (less than 5% in total) suggests that consolidation is at a low level. The high level of empty running (consistently around 26%) supports this and suggests that back loading is still fairly limited. About 90% of kilometres are run with one destination per journey.
3. Long distance vehicles used for local deliveries
A third explanation is in some ways the opposite of the second, that HGVs are being used for local deliveries to avoid transferring bulk loads into smaller lorries. Thus the lorries run from a central depot with a consolidated load to the edge of a town and then undertake a number of deliveries using the local network. The high cost of using a large vehicle is offset by avoiding the transfer cost to a smaller vehicle. This is related to how much of the network is open to use by heavy vehicles. At the moment there are relatively few restrictions in the UK.
The tabulations from CSRGT show some evidence supporting a small percentage of total articulated HGV traffic (around 5%) undertaking 6 or more drops per journey. This suggests a small number being used for multi-drops beyond what would be expected in the case of consolidation of medium sized loads.
Table A1: Proportion of distance travelled by artics over 33tonnes maximum vehicle weight on single and multi-drop journeys
Combining the distribution of loads in Table A1 with the data in the charts suggests that the “vehicle doesn’t match load” hypothesis is supported, although not proven, by the data. More detailed research is needed to test this further, but there is also prima facie evidence that consolidation is at a low level, whileit is clear that empty running is high.
Overall, a move to greater consolidation, and better matching of vehicle size to the work being demanded of them would provide a sensible way forward. As well as proper charging for external costs, for foreign registered vehicles as well as UK operators, this should be borne in mind in the design of the proposed Lorry Road User Charging (LRUC) scheme.
1. See Transport Statistics Great Britain 1996, 2006 and 2009