Back to Home Page                                                                                                                                                                                                                         

12 June 2001

Railways came into existence nearly 200 years ago after poor road surfaces were first improved by laying timber or iron plates, on which horses could haul greater loads. Since the Second World War, proposals have been made to reverse the process and convert railways into roads. The most recent support for this old chestnut appeared in the Sunday Telegraph on 3rd December 2000. Dominic Lawson wrote of railways as “straight lines”, a description that is not borne out by an examination of the infrastructure. It is an historical fact that many landowners compelled railway companies to follow circuitous routes. 


“The Twilight of the Railways”

The first conversion proposal [1] was advanced by Brigadier T.I. Lloyd in the 1950’s. He envisaged:-

     A reserved toll road on the converted railways, with no dangerous bends, having single carriageways which were safer than dual carriageways because drivers would have to remain alert! 

     That use would be restricted to vehicles and drivers, both of which would be subject to stringent tests.

     High professional standards of driving would be instituted, and subject to special driving licences. 

     Traffic would travel at an average of 60 mph, at 100 yard intervals, users achieving that voluntarily over the entire system, round the clock and over the four seasons.

     If drivers slowed down excessively at 24 ft bridges and tunnels, the question of widening should be taken up.

     Prompt sanding and snow clearance, and radar fitted to vehicles to operate safely in fog. 

     New traffic rules: keep strictly in lane, except at clearly marked places of transfer to adjacent lanes; no overtaking at all on two lane stretches of roadway; no dawdling; special driving licences would be forfeited for breaking the rules. 

     Buses, with 40 seats, would have one or two trailers, which could be discarded in the non-rush hours and could also be used as slip coaches. The buses would run when full, not to a timetable.

     There would be no problem with the rush hour at Waterloo, the gross daily passengers are no more than 200,000 - requiring 60 seater bus-trailer combinations from each of the 21 platforms at the rate of one every one & a half minutes. He claimed that the rush hour flow of 840 vehicles - one every 4.2 seconds - would not amount to the full capacity of one single lane and mentioned the multiplicity of lanes at Clapham Junction. 

     Single track railways would be readily converted into sub-standard two lane roads. The expense of raising them to full 22 ft standard would no doubt be justified in some places.

     Existing rail passenger and freight traffic would require only 10,300 vehicles at 60 mph. These would consist of one-third forty-seater buses and two-thirds twenty-ton lorries. These powered vehicles would be supplemented by an abundance of trailers.


Debated by the Institution of Civil Engineers

His proposal was demolished by road experts who attended the debate [2] in November 1955:-

Dr. Glanville, Director of Road Research, Department of Scientific & Industrial Research - forerunner of the Transport Research Laboratory: “Could not see how buses could run, fully loaded for eight hours a day, six days a week, and if this did not happen, the financial basis for conversion was affected most seriously.  The Minister of Transport had stated that traffic lanes on motorways would be twelve foot wide, further affecting the proposals which were based on ten foot lanes”.  Dr. Glanville did not accept, that “higher standards of driving, would be sufficient to overcome the dangers of high speed traffic on the same carriageway”.

Major Aldington, Technical Adviser, British Road Federation [3]: “Knew no one who believed that a single 22ft or 24ft carriageway was adequate for heavy volumes of traffic and said that it was quite preposterous”.  He “viewed with alarm the prospect of travelling at 60 mph on single carriageways against opposing traffic, particularly at night with glaring headlights.  Driving from London to Birmingham, [in 1955], at night was quite terrifying.  It was impossible to get a 20ft carriageway through an ordinary double line rail tunnel”.

Mr. Burnell, London Transport Buses: “It was totally impracticable to ask staff to drive on a road 22-24ft wide at such speeds. They would rightly refuse”. He postulated a driver handing over to a relief driver saying: “Engine pulling well, there is ice and snow and fog, but the radar is all right”.

Mr. Osborne, Resident Engineer, Wilson & Mason: “Highway width must be 88-93ft with dual 22ft carriageways and 15ft verges. Double track rail formation is 39ft reduced to 19-22 ft in tunnels and deep cuttings with retaining walls; quadruple track was only 55ft wide. Gradients are a serious problem; less than 1 in 200 is inadequate for water drainage on roads, hence a completely new drainage system would be necessary”. He contradicted Brigadier Lloyd who had said there were no dangerous bends, pointing out that “Railways had curves of 660-1,320ft radius, often on viaducts, bridges or through tunnels, against a motorway standard of 2,865ft and said conversion had no potential and was economically impossible”. He added that “railway services would be completely withdrawn long before roads were laid on which replacement services would operate”. 


Other objections included:

     Most railway junctions are on the flat and would have to be replaced with flyovers or clover leaf junctions to allow traffic to keep moving. 

     The enormous number of bridges and tunnels - 63,000 and 1,050 respectively [4] - represented major engineering obstacles; 30,000 under bridges would require new floors because rails were laid on longitudinal girders which could not carry a roadway,

     Much railway was built on embankments or in cuttings - enormous road works would be required.

     The incidence of accidents would be bound to increase.

     No vehicular provision would be available for seasonal peaks, nor for sporting events.


Conversion was supported by Professor Bondi, Kings College, London: “A twin tracked railway was wide enough for one carriageway, without a parking strip. A ten foot lane was adequate, more than eleven was positively dangerous in encouraging small cars to pass. The Marylebone-Sheffield line could be made a northbound road, and the Sheffield-St. Pancras a southbound road!”


Other Weaknesses & Problem Areas

Surprisingly, no attention was focused, in the debate, on level crossings. Brigadier Lloyd claimed that “junctions and level crossings were relatively infrequent”. There were 24,000 level crossings over the 20,000 route miles, plus thousands of junctions. He had not said how many crossings there were, although the statistic was publicly available. In addition to 4,670 public level crossings, there were 19,700 unmanned Accommodation and Occupation crossings providing access to farmer’s fields or residential property, plus 2,500 public footpath crossings [5]. The attendant dangers and delays from tractors and animals on crossings would be a serious problem, which he had completely overlooked. The opportunity for them to cross between vehicles travelling 100 yards apart at 60 mph would be virtually zero. New bridges would be necessary at all such points. Nor did anyone challenge the prospect of 60 mph buses slipping trailers without stopping on a single carriageway - with another vehicle 100 yards behind. 

The cost of his “mobile police and efficient breakdown service” was not brought into the equation.  Indeed many costs were not brought into the equation

On 16th February 1955 the Minister of Transport informed Parliament [6] that, except in a few instances, it is prohibitive to convert redundant railways into roads. 

A third of route mileage was single track, and a further 50% was only double track:-


No. of tracks

route miles [7]

Single line


Double line


Triple line


Four or more lines




Track miles





An examination of the Department of Transport publication  [8] on railway construction standards would have shown him that track widths were too narrow for use as roads. Moreover, many rail routes were below even those prescribed widths for historical reasons. 


The British Road Federation “Roadshow”

In 1984, the British Road Federation held a series of meetings around the UK to pursue a campaign for improved roads. In Manchester, British Rail were invited to attend and I represented them. The speaker was Mr. Gent, Director of the Federation.

The speaker suggested converting under utilised or unused railways into roads. His views were in conflict with those expressed by his predecessor at the Institute of Civil Engineers debate in 1955. 

     I said that only sixty out of 7,000 miles of track closed in the past twenty years had been converted into roads and before they set their sights on lines that they thought were under utilised, they should convert the 6,940 miles of closed routes begging to be used. 

     On technicalities of conversion others present had no facts. I pointed out that an independent survey by Coopers & Lybrand Associates [9] showed that only one of ten lines recently considered for closure had any prospect of conversion. Even that line (Marylebone-Northolt), had less width than the minimum 7.3 metres required by the Department of Transport for carriageways: being 6.9 metres or less, with 5.9 metres in one place. 

     I drew attention to the cover of their publicity booklet depicting a High Speed Train on a single line passing under a hump backed single arch bridge of low clearance, pointing out that it exposed the conversion problem very clearly! 


The speaker also disregarded the costs of emergency services [10],  the effects of pollution, structural damage to buildings, pavements and verges and double glazing all of which fall on others.  Contrary to industrial opinion, road haulage pays much less than it should for road use. Recent research shows that “LGV’s only pay for around 59-69% of the costs they impose on society.  Per tonne carried, rail produces around 80% less carbon dioxide than road”. [11]  


The road lobby argues that road taxes should be spent on roads. During the 123 years when railways were privately owned, they were subject to corporate and other taxes, not a penny of which was returned to them to spend on their “highway”, built entirely at company expense. A further tax imposed on railways, from 1830 to 1929, but never on trams or buses, was “Passenger Duty” [12]. During two wars, Government froze rail prices at pre-war levels, to hold down Government expenditure by over £1bn, in addition to imposing taxation and Excess Profits Tax, whilst road transport profits were untouched, and even allowed to escalate [13].  Unlike all other businesses, railways were not refunded 20% of Excess Profits Tax after the 1939-45 war. British Railways did not pay taxes because Government interference held fares up to 41 points below the inflation rate for 34 consecutive years, thereby cutting revenue by some £8billion [14]. The road lobby overlooks that users get a hidden refund - from savings in running costs gained by using new roads. 


The Railway Conversion Campaign

In a 1970 booklet [15], the conversion lobby, list 29 instances of closed railways in England, Scotland & Wales, totalling 43 miles, that had been converted into roads, including five single lines converted into dual carriageways. Fifteen were less than one mile in length. The average length was 1.5 miles. Only five of the cases were double lines. The rate of conversion was less than five miles pa.

The road lobby places great emphasis on the subsidy paid to keep open rural branch lines, that British Rail were prevented from closing by the Government! The fact that 15 years later, only 0.8% of closed lines which ranged from about 20 to 180 miles in length, had been converted to roads, serves to show that railway lines are not suitable for conversion. The road lobby average out total rail traffic over the full route mileage, ignoring that two thirds of the system carried 99% of rail traffic [16].

In 1989, the Conversion lobby regurgitated Brigadier Lloyd’s discredited theory, and advertised in the media [17]: “The railway system is only working at 3% of its potential”,  “a Department with such poor utilisation ought to be sacked” and expressed concern for “our precious green land”.  My response to the Daily Telegraph pointed out:-

     Road utilisation was worse than rail, having 22 times as much road mileage, some 80-100 times as much acreage, for ten times as much traffic. 

     If the League is concerned about “our precious green land”, 60% of roads should be closed to bring road utilisation up to British Rail’s level. 

     Under-utilised lines are mainly in rural areas, and were kept open by Government decision, without subsidy for the first twenty years of nationalisation. British Rail had to fund them from interest bearing Treasury loans, which together with fares held below the R.P.I. had caused the crippling deficit [18].

     To be proved fatally wrong, anyone believing that rail utilisation is 3%, need only sit on a main line for a few minutes, not 58 minutes in an hour, which is the 97% that they claim is unused. 

     6,940 miles of railway closed since the 1960’s was available for conversion but remained unused .

     Roads are built on the basis of social benefit, an ingenuous formula based on the time road users may save by using new roads. British Rail in contrast, had to justify investment in money terms. 


The Problem of Road Utilisation

A principal reason for poor utilisation of roads is speed variation: 30-100mph on motorways, 10-70mph on other roads. It is bad management in commuter peaks, to permit horses, 10 mph tractors and lorries so heavily loaded that they cannot exceed 20-30 mph on a 60 mph road. Delays at all times of the day, are caused by very wide loads that make their majestic progress at horse drawn speeds. Some, unsupervised, technically legal, wide loads are encountered on minor roads causing delays and potential danger for other road users. 

The road lobby ignores the fact that juggernauts have to leave motorways to deliver in towns and villages, where they negotiate road junctions by halting traffic movement in all four directions. Some commercial premises have such restricted access that vehicles shunt to and fro for up to five minutes to effect entry. The ensuing delays can only be reduced by gutting commercial premises and reconstructing local roads, with its attendant compulsory purchase of property. More delays are caused by using roads to unload car transporters and other vehicles for firms with inadequate access to premises. Selfish conduct, bad lane discipline, and failures to observe the Highway Code exacerbate the problem. 

Despite the appalling under utilisation of roads, users create worse problems. Ultra brief media reports inform of mind-boggling hold-ups due to lorries turning over, jack-knifing, shedding their loads and losing tyres, and many other vehicles travelling too fast and too close. Ensuing delays are never translated into time or money loss. No compensation is paid by offenders - if it were, via a national fund, delays would plummet. 

Converting railways serving small towns and villages into roads would not change this situation. The tracks, being double or single, would not offer the huge turning areas required for LGV’s.

Hitherto, it has been claimed that heavier goods vehicles reduce the cost of goods to the consumer. No account has been taken of the increased costs borne by motorists who are delayed behind heavy vehicles on “A”, “B” and other roads. Most of this road mileage is not dual carriageway and is on routes that are not paralleled by railways, which the conversion lobby argues, could be converted to roads. Indeed, if 11,000 miles of railway were converted to roads, it would not benefit one jot, millions of delayed journeys on 220,000 miles of roads remote from a railway.

After 40 years of motorways, no system has been developed which effectively diverts traffic held up by an accident onto alternative routes. My experience is that radio warnings come after one has passed the last exit, and arrived behind a two hour queue. Additional alternative routes such as railways converted into roads would therefore be wasted.      


The Fall-out from Conversion

Converting railways into roads would produce a serious Doomsday scenario, creating an unending and inescapable burden, far in excess of any rail subsidy. An inability of the Exchequer to swallow the increasing burden of road costs would focus economies on other Public expenditure. Bridge costs, hidden in railway “losses”, would fall directly onto the roads budget, where they should have been since 1930, as the Minister of Transport had recommended [19]. Overbridges would have to be widened and lifted to reduce the risk of bridge bashing. 

The burden on the State would be exacerbated by the consequential increase in fatalities and injuries. In a letter to the Daily Telegraph in 1991, I pointed out that railway accidents from 1952 - which the Telegraph [20] had used as a base line to criticise railways, had killed an average of eight p.a. compared to an average of nearly 5,000 p.a. on roads. The number injured by accidents in both modes is equally disparate. The cost of this excessive number of road casualties is a hidden cost of road transport. Given that there was about ten times as much traffic on roads as railways in 1991, conversion would have increased fatalities from eight p.a. of those using rail to one-tenth of total road fatalities - about 500 p.a. A letter in the Sunday Telegraph on 31st December from Andrew Dow said that in last 75 years, 442,000 people were killed on the roads and 19 million injured.  Whilst fatalities have decreased on roads, they have decreased likewise on railways, hence the relative disparity remains. “Lorries were responsible for a fifth of all road deaths last year” [21].

The road lobby argues that road casualties are greater due to the mix of pedestrians and vehicles. This is easily resolved - at the expense of those who began to use roads after pedestrians, viz. mechanised road transport. Roads could be fenced along their entire length as railways were statutorily compelled to do, with controlled gaps at selected places.   


New Messiahs

A second Sunday Telegraph article [22], highlighted longer gaps between trains as compared to road vehicles.  Such a comparison is meaningless. A passenger train carries as much as 300 cars [23], at almost twice their speed. As the safe headroom between road vehicles is one metre for every one mph [24] the aggregate headroom for 300 cars, at 70 mph, on one track/lane, would be 21,000 metres, ten times the “long gap” maintained, for safety reasons, ahead of a high speed passenger train [25]. If cars travelled at train speeds, their aggregate headroom would need to equate to 15 times the “long gap” between trains.  A freight train can convey as much as 200 lorries [26], and travels 25% faster. Travelling at 60 mph, at one metre for every one mph, they require an aggregate 12,000 metres of headroom, seven times as much as a 75 mph  freightliner [27]. At 75 mph, 200 vehicles would require 15,000 metres aggregate headroom - ten times as much as one freightliner requires.  Figure 1 illustrates the relative headroom disparities:-



Figure 1


This Sunday Telegraph article quoted Sir Phillip Goodhart who had observed a stretch of the M6 close to the main London to the North West railway that “was jammed solid while the rail line which was slightly wider than the M6, had almost no traffic on it”. Having been responsible for operating the whole of that route, and as a user of the M6, I can say that such a location is not typical. Much congestion on roads is caused by bad driving and the ensuing accidents. The width required for a motorway, with three lanes each side, is 35.6 metres [28]. The overall width - on a comparable basis - required for a four track railway is 16.5 metres and 8.08 metres for a two track route [29]. The West Coast line has two tracks for about half of its length. The railway route is wider at stations, but these must be compared to service areas that occupy far more land. In addition, those huge clover leaf junctions on motorways, especially on the M1/M6 are in a league on their own. The relative requirements for motorways and railways are depicted in Figure 2. The outer box represents the width of land and bridge height required for a three lane motorway; the larger hatched box is the comparative width and bridge height required for a four track railway; and the smaller hatched box that for a double track railway. It will be seen that the minimum width required for a three lane motorway is more than wide enough for two major main lines!



Figure 2


Converting Roads into Railways

In view of poor road utilisation, a better option may be to convert some roads into railways. The annual goods vehicle km of 10.8 billion on motorways [30] equates to only 185 vehicles per hour in each direction.  Given the average load of goods vehicles is 5 tons, it represents less than one decent train load, which would not tax one road lane converted into a railway. Huge benefits would accrue. Motorway construction and repair costs would plummet, and contraflows would be much less frequent, leading to higher average car speeds. Total fuel consumption would fall and fewer lorries would be imported, leading to an improved balance of payments. Fewer accidents would reduce costs of the emergency services and the Health Service, whose waiting lists would fall in consequence. Lay-by’s used as cost free premises for overnight or weekend lorry parking would be unnecessary. 

“Heavy lorry mileage on journeys over 150 km represents 50% of all mileage and 20% of all goods.   Transferring this to rail would cut total lorry mileage by a half” [31]. If trunk haulage was by rail, using smaller containers, haulage into towns and villages on shorter vehicles would slash congestion, and end damage to pavements and verges - some of which compare unfavourably with ploughed fields.

Converting some roads into railways would not face the same transitional problems as the converse proposal. Firstly, random observations indicate that there is significant capacity on existing freightliner trains.  Secondly, the under-utilisation of roads, to which I referred earlier, allows scope for re-routing road traffic during the changeover period. Much cross empty mileage would be eliminated by central control of containers.  


Top of page                                                                   For more see “Railway Conversion – the impractical dream”


[1] Later published as the “Twilight of the railways ”, by T.I. Lloyd (Forest Groom & Co.)

[2] Proceedings of the Institution of Civil Engineers, November 1955:  “Potentialities of the British Railways system as a reserved roadway system”

[3] The road lobby

[4] British Railways Year Book 1963

[5] Report to the Minister of Transport on the Safety Record of British Railways for 1957 by the Minister’s Railway Inspectors

[6] Hansard, vol. 437, col. 47

[7] British Transport Commission Report & Accounts 1956

[8] Department of Transport “Railway Construction & Operation Requirements”

[9] British Rail Internal publication “Management Brief”, 15th March 1984

[10] A 1994 “Police Stop” video stated that road accidents cost £5.5bn p.a.

[11] “Goods without the Bads”, Transport 2000 Booklet

[12] “Square Deal Denied”, ISBN 0-9521039-3-1, page 18

[13] ibid, Chapters 12 & 13

[14] “Blueprints for Bankruptcy”, ISBN 0-9521039-2-3, page 63

[15] “Conversion of Railways into Roads”, ISBN 0-9046980-1-7

[16] British Railways Board Report & Accounts, 1963, para 4

[17] Advertisement in The Daily Telegraph, 26th July 1989

[18] “The Railway Closure Controversy”, ISBN 0-9521039-4-X

[19] “Square Deal Denied”, page 64

[20] The Daily Telegraph, 9th January 1991

[21] “Goods without the Bads”, op cit

[22] The Sunday Telegraph, 24th December 2000

[23] Average load of cars = 1.6 [Annual Abstract of Statistics, 1998, Tables 14.2 & 14.4]

[24] Highway Code & LGV School Instructors

[25] Railway Signalling Standards

[26] Average load = 5 tons [Annual Abstract of Statistics, 1998, Tables 14.1 & 14.4]

[27] Railway Signalling Standards

[28] Highways Agency

[29] Department of Transport “Railway Construction & Operation Requirements”

[30] “Transport Statistics of Great Britain”, Table 4.9

[31] “Goods without the Bads”, op cit