Biting the Bullet

What we can learn from the Shinkansen

Christopher P. Hood, Director, Cardiff Japanese Studies Centre, Cardiff University [About | Email]

Volume 1, Issue 1 (Discussion Paper 3 in 2001). First published in ejcjs on 23 May 2001.

This paper is the start of a new research project for me, having just completed a six year study of the reform of the Japanese education system and the influence of former Prime Minister Nakasone over that process. However, this research has been 13 years in the making as I have continued to read about, as well as use, the shinkansen over all the years that I have been going to Japan. Central to this paper is the research conducted in January 2001 when I spent three days with Central Japan Railway Company (hereafter referred to as JR Tokai, as it is better known in Japan). I would like to express my thanks to all those who helped in that trip, in particular the International Office in Tokyo and Mr Kondoh at the London Office of JR Tokai.

What I will be focusing on in this paper is only one part of my research on the shinkansen. My research will be looking at the development of theshinkansen, the way that it has become a symbol of modern Japan, and what Britain (and other countries) can learn from the shinkansen. This paper will concentrate on the last of these points.

I have been very pleased so far to discover that there is a lot of interest in my research on the shinkansen. However, at the same time, this enthusiasm fills me with some sadness as I am all too aware that part of this interest is a result of the problems with and recent disasters on the British railways. The year 2000 was a bad year for British railways, indeed for the symbols of British transportation as a whole, what with the crash of Concorde too and the withdrawal of the hovercraft service between Dover and Calais.

A survey recently asked British people what they believed to be the cause of the problems on our railways, with possible answers being the Conservative government of Prime Minister John Major (for the way that they privatised the railway industry), the present Labour government of Prime Minister Tony Blair, or the railway industry itself. The results of that survey are not significant for this paper, but the question itself reminds me of the teasing question I saw on television which asked 'What caused the Titanic to sink?', with the possible answers being an iceberg, the captain or the design of the ship. Although it is often preferable, and certainly popular, to single out one cause, in reality it may be a combination of different factors. It is important at this time not only to understand how we have reached the present situation but, perhaps more importantly, we need to start looking to the future and thinking about how we can make improvements. It is here that the shinkansen becomes significant for Britain.

Why Study the Shinkansen?

The shinkansen is the world's leading high-speed inter-city train service. The key points about it, which will be looked at in more detail below, are its excellent safety record, its high speed and its punctuality. The shinkansen, as a result, has enjoyed extremely good press in Britain and other countries and, although mistakes are often made in the popular media when reporting or showing the shinkansen, they tend to be positive ones.1 It is important to understand that the shinkansen is not perfect. There is room for improvement still, and the Japanese railway companies and related research institutes are constantly looking to make further improvements. In fact, I would even suggest that there may be some improvements that could be made by learning from British railways and trains (for example, in relation to reserving seats, luggage space on trains, and the designation of smoking areas). It is also important to be aware that the shinkansen sits at the very top of the hierarchy of Japanese trains, and in many respects is not directly comparable to British inter-city services, which often better resemble tokkyu (limited express) services on ordinary lines. However, the way that it is run clearly reflects and has an influence on the way that other lines in Japan are run, and so there is much to be looked at and possibly learnt from it.

The shinkansen started in 1964, coinciding with the hosting of the Olympic Games in Tokyo. These two events symbolise the start of modern Japan. As those who regularly watch 'Who Wants to be a Millionaire?' on television will know, the shinkansen is in fact the name of the line rather than the train, though the word has come to be used for both (and more commonly the train). The term 'bullet train', which I do not like using myself (despite using it in the title of this paper), dates back to the Pacific War when a 'Bullet train' (dangan ressha) was planned to speed up the link between Tokyo and the Asian mainland with a dedicated track being built as far as Shimonoseki. Indeed, work started on this plan, and although the project was abandoned as the war turned sour for Japan, some of the tunnels were made and much of the land was purchased, which made the eventual construction of theshinkansen much easier. However, due to the length of the word 'shinkansen', as well as the shape of the original model, 'bullet train' has become the standard term used in many English-speaking countries.

When the service started in 1964 overnight it cut the travel time between Tokyo and Osaka from six and a half to four hours. Today it can be done in a mere two and half hours. Over the years the network has also been expanded with the Tokaido line being extended beyond Osaka to Hakata in Fukuoka (Sanyo line). Additionally, lines going north from Tokyo to Morioka, Niigata and Nagano have been added. On top of this there are presently two 'mini-shinkansen' that operate between Morioka and Akita, and between Shinjo (in Yamagata) and Fukushima; both with through services to Tokyo. The difference between 'normal' shinkansen and 'mini-shinkansen' is that the 'mini-shinkansen' are not on dedicated lines. Although the lines have had to be widened to take shinkansen (the standard Japanese gauge is narrower than the shinkansen gauge), there are also 'local' trains using these tracks, as well as there being level crossings, for example.

Shinkansen - High Speed

The shinkansen does not push technology to its limits in the pursuit of speed - unlike Formula 1 racing, for example, where this often leads to blown engines. It is the areas of continued environmental improvements, safety and efficiency that have been the greatest concern. Having said this, the 500 series shinkansen (operated by JR West) is officially the fastest train in the world as it has the highest average speed between stopping stations, with a top speed of 300kph and an average of 261.8kph between Hiroshima and Kokura. In contrast, Britain's best is provided by the GNER service on the East Coast main-line with a top speed of 200kph and an average of 180.2kph between London and York.2

Figure 1. A 500 series passes a 0 series

Hood, Figure 1

What is particularly significant about the shinkansen in this area is that they run near to top speed for much of their journey as there is a lack of corners, inclines, and level crossings, and, moreover, they run on a dedicated line (and so do not have to worry about other trains crossing over or joining the line). This allows the companies to run trains more frequently, and makes it easier to run them on time and safely. The end result is trains often departing as regularly as every five minutes on the Tokaido lines, with the 400 metre long trains often full with over 1,000 passengers on board.

In Japan if a train is more than 60 seconds late it is officially declared to be so (compared to five minutes in Britain). The average delay on the shinkansenis 0.4 minutes. This figure has recently risen to 0.7 due to some exceptionally strong typhoons in Autumn 2000 (needless-to-say, the high frequency of punctual services helps to ensure that this average does not rise too significantly). In the control centre at Tokyo station, there is a display of all theshinkansen running at any particular moment, with the length of delay displayed next to each train. On the day that I visited, every single one read '0'. On top of this, when I travelled in the cab of a shinkansen the next day, as we passed through stations the display read that we were zero seconds behind schedule, though at other times we were as much as fifteen seconds out! It is also important to note that the trains never arrive early - something that does happen from time-to-time in Britain (which the conductor or 'train manager' often 'proudly' likes to announce), which raises the question why it cannot always arrive at that time.

Figure 2. A shinkansen on time ... to the second!

Hood, Figure 2

Shinkansen - Safety

Since the shinkansen started there have been no fatalities on the service due to collisions, derailments and so on. I have heard a story of someone dieing from a heart-attack due to the shock of the high speed (though have yet to find evidence to support this story), and one child was tragically killed after being trapped in the door at Mishima station. As I stressed before, the system is not perfect. Accidents do occur on normal lines, but these are comparatively rare (though accidents at level crossings remains a major concern). There have been accidents on the 'mini-shinkansen' - for example, one man died in January 2001 after driving his car up the track on the Yamagata shinkansen having apparently become disoriented in the snow at a level crossing.

What is interesting about the shinkansen - and is clearest to see from a foreign perspective - is the lack of signals on the line. Instead a system known as ATC is operated. Using the 'block system' developed in the early days of rail in Britain, it limits the speed a train can go based on where the nearest train in front is. The more empty blocks, the higher the speed allowed.

However, perhaps more important is not this system (and a new improved one is presently been developed), but that the drivers stick to the rules. The level of training is extremely high and is, of course, based on the foundation of an extremely disciplined (despite the problems discussed in recent years, it remains so) education system. It is in this area that Britain almost certainly will have to pay much greater attention if future disasters are to be avoided. Some slips have occurred in Japan - for example in February a driver left the train running while he went back to retrieve his cap which he had left in the wash area of another carriage (the incident was discovered as he was questioned about why the train arrived one minute late at Tokyo station) - but even then the ATC system and other procedures ensures that no-one is any physical danger. It is perhaps a shame that the Japanese nuclear industry does not appear to be as well run as the shinkansen in this respect.

Another feature of the safety - and perhaps one that would be a problem for Britain if it wanted to run a similar system - is that services end before midnight each day and do not resume until six. The reason for this is to allow maintenance work to be carried out daily as required (as opposed to running slower or detoured services on Sundays or simply closing lines when essential work becomes necessary, as in Britain). On top of this Dr Yellow - a special train that checks all aspects of the lines and overhead power cables - regularly runs along the lines (at normal shinkansen operating speeds) in order to check for any potential problems. Again I was fortunate enough to travel on Dr Yellow,3 and was even there to experience the moment when a fault with a small section of track was discovered. Dr Yellow in many respects is the symbol of the safeness of the shinkansen - although ironically, most Japanese I have spoken to (other than young boys - and their mothers (who tend to be experts on all forms of transport) are unaware of who or what Dr Yellow is.

Figure 3. The old Doctor Yellow (retired in Winter 2001) arriving at Nagoya Station

Hood, Figure 3

Is the Shinkansen Peculiarly Japanese?

Japan is a long, narrow country. This makes it particularly suitable for developing a rail network - though its abundance of rivers and mountains obviously is less of advantage. However, much of the Japanese population is located on the Eastern seaboard, and so can easily be connected by one line. This is opposed to the multitude of lines that are needed in Britain, though high-speed lines could be developed that would have better population coverage than now.

Ironically there are in fact more railways (per head of population (279.8 km/million people in UK compared to 142.2 in Japan), and in relation to the area of the country (67.8 km/100km2 in Britain compared to 47.6 in Japan) in Britain than in Japan. However, British people do not use them as much (6% compared to 35% of journeys) - although the fact that so many British are so outspoken (especially in criticising the lines or services) may give the impression than more people use them.4

The high level of education and professionalism of the shinkansen operators and drivers need not be a problem in Britain, though it would doubtless take time to change the bad-habits that have managed to fester and develop over the past few decades (and if the level of service at McDonald's, for example, is anything to go by - where it is still nothing like that in the United States - then perhaps this will be an almost impossible hurdle). In other words, as one JR Tokai employee told me, 'you cannot give other countries the shinkansen and expect it to work perfectly with no accidents'.5

Successful Privatisation

Japan National Railways (JNR) was privatised in 1987. Central to the reform was trying to ensure that the companies could be financially viable. Therefore, the three main companies on the main island of Honshu were each given a part of the profitable shinkansen network that could be used to cross-subsidise the less profitable, and often rural, lines that were also to become part of their franchise. Japan was therefore divided up into six regional blocks - East, Central and West on Honshu, then also Hokkaido, Shikoku and Kykshu. On top of this, one freight company was also created. It should be remembered at this point that there are numerous fully private regional railway companies as well, that remain wholly separate from Japan Railways (JR). A key point to note - in comparison to the situation in Britain - is that the operating companies also own the lines, trains, stations and so on. This ensures total responsibility is taken by one company and that intermediary companies are not each trying to make a profit, which translates itself into high ticket prices. On top of this, it should also be noted that companies have been looking to make money outside the 'normal' channels of the rail industry - most notably JR Tokai's ownership of the 'Matsuri' Japanese restaurant in London.

Although the government continues to have a sizable holding in many of these companies, the 'privatisation' of JNR is seen by some as the most significant moment in Japan's railway history. Although the companies are still burdened with huge debts from the JNR days, the Honshu companies at least are now profitable and it cannot be doubted that the attitude and performance of the rail industry as a whole has improved greatly. As far as theshinkansen is concerned, since privatisation we have seen the development of new models and the continued expansion of the shinkansen network.

I do not want to go into detail about the privatisation of the railways in Britain, but it is worth noting some of the key points here for comparison. One of the central problems (if we are to assume that the Japanese privatisation provides a successful model) is that Britain broke up the network into many franchises. Not only are there too many of these franchises - with some having to run rural (unprofitable) lines - but the franchises themselves are often too short, which discourages investment in both the railway lines and the rolling stock. Second, the division of stations and lines, rolling stock, and franchises (combined with the nature of British geography and the many franchises that lead to the need to share certain stretches of the network) results in difficulties with timetabling, costing and responsibility (further compounded by the strange complications of there being the 'wrong snow' or 'leaves' on the line!).

The Way Forward - The Japanese Model

Following the crash at Hatfield in 2000, there has been a huge maintenance programme underway to replace worn out track. Although such action is commendable - though obviously well overdue - it appears to be creating a situation whereby some sections of the public are expecting this to solve the problems on Britain's railways. It will not. Therefore, if we are to continue with the assumption that Japan - and particularly the shinkansen - provides a useful model from which Britain (and other countries) can learn, what are the key points that need to be taken into account and then made into policy?

The first thing that would appear to be clear is that the idea of re-nationalisation would be foolish. Although some would suggest this would help take us back to the 'good old days of BR', I am unsure as to what 'good old days' such people are referring to. Certainly Britain led the way during the great age of steam between the wars, but there can hardly be anything good to say about the state of British railways through much of the 1960s, 70s and 80s. In fact they could be described as a national embarrassment. It is time to look forward, not back (though of course the steam days were also one of private enterprise). However, in some areas government intervention and re-nationalisation may be a stepping stone to the eventual result of an improved form of privatisation.

The first suggestion, therefore, would be to get rid of Railtrack, which owns and operates the tracks and stations, among other things. Although there may have been much good with the original theory of such an organisation - it is clear that it has not worked well with the present system. Leaving franchises relying essentially on passenger revenue alone, from the 'Japanese model', would also appear to be a mistake.

Second, the length of the franchises needs to be dramatically increased (if we are not to create fully private companies). I would even suggest increasing them to as much as 50 years, with the provision that if they do not meet certain agreed targets (cost, punctuality, etc.), then the franchise can be withdrawn with one year's notice from the rail regulator.

Next, it would appear logical also to decrease the number of franchises on the British mainland to just five, four of which (like Japan) would have a major inter-city route, from which money could also be used to help maintain the lesser used lines. However, like Japan, some lines could also be given to the 'third sector' - to be run by local companies or organisations independent of the main network (though through-ticketing is still possible to some degree).

The four main franchises would be; East Coast and Anglia (based in Kings Cross, St. Pancras and Liverpool Street stations in London), West Coast and North Wales (based in London Euston), West and South Wales (based in London Paddington), and South Coast (based in London Waterloo and London Victoria). The fifth franchise would be freight (an area, it should be noted, where Britain has been considerably more successful than Japan).

On top of creating these new four super-franchises, other reforms should also be implemented. First, make the system easy to understand and enjoyable to use - this will help win back the car (and plane) users. In other words, as much as possible, the complex system of ticketing (which presently includes standard, saver, super-saver, Apex, Super-Apex, Advanced, Day returns, etc.) should be simplified. Supplements (as has already effectively started on some lines) should be needed for using high speed inter-city services compared to 'local' trains. Also, improvements should be made to the seat reservation system (including perhaps a surcharge). Neither Britain nor Japan has an ideal system, but with modern technology a better one must now be possible. Seats should be more comfortable - and even reversible, for although the tables on British trains are pleasant, there are many who do not like having to travel backwards in many of the other seats in carriages.

In some areas, we have already seen some improvements and the above recommendations are likely to further enhance these. For example, the infamous lack of announcements on British trains has now be dealt with - though for some reason, it still appears to be impossible for the announcements to give you the really useful information such as which platform and what time a connecting service departs from, for example. We have already begun to see some franchises, such as Virgin, investing in new and improved rolling stock. Also, naming the trains, by which I mean the service rather than the vehicle itself, which at present tends to be limited to just the fastest services (particularly in the morning) in Britain, should also be encouraged to help ensure passengers catch the right train (naming of trains in Japan has also helped lead to greater connection between people in certain regions and their main service, with naming often done through voting and suggestions from the public).

The next suggestion is that these four super-franchises begin to work towards developing some new high-speed lines. Although all the suggestions so far should help lead to improved services, and theoretically the present network could be sped up further and made safer, one should be in no doubt that new dedicated lines would be a major advantage. Central to these new lines would be that they would reduce the strain on the present lines and so free up capacity for when demand increases (as well as making it easier to put on extra local and freight services, which not have to worry about 'blocking' faster trains), the services will be fast and punctual (as there will be little, if any, need for lines to cross or for there to be level crossings, and they could be straighter and electrified, for example), and that they should not be restricted to the present routing, but aim to cover populated areas better. Such investment will lead to better usage of public transport (after all travelling by car is often faster and cheaper for inter-city journeys than going by train) as well as regional regeneration if we are to assume the results would be similar to that of Japan.

The four super-franchises therefore could have the following lines:

  • East Coast
  • London - Luton - Northampton - Leicester - Derby - Nottingham - Sheffield - Leeds - York - Middlesbrough - Newcastle - Edinburgh - Dundee - Aberdeen
  • West Coast
  • London - Oxford - Coventry - Birmingham - Wolverhampton - Stoke - Manchester - Liverpool - Preston - Carlisle - Glasgow.
  • West & South Wales
  • London - Reading - Newbury - Swindon - Bath - Bristol - Cardiff - Swansea.
  • South Coast
  • London - Gatwick - Brighton - Portsmouth - Southampton - Bournemouth/Poole - Exeter - Torquay - Plymouth.

As in Japan, government funding will be essential. However, if done properly, there is no reason why these investments should not be viable. Although it is true to say that the JR companies still have much debt, the newer lines have proven to be profitable - and much of the debt is due to the way JNR was run rather than the development of the shinkansen. However, without a substantial change in attitude from the present way investment and the British rail industry is viewed in the City, government support will be needed.

Another potential problem is that in some cities - for example, probably Sheffield - limited space may mean that new special stations for the high-speed lines may be needed. This can be seen in Japan too - most notably in Osaka, with the shinkansen serving Shin-Osaka station rather than Osaka station itself. However, this problem should be seen as something that will help urban and regional regeneration (as well as local transport) rather than as being a major hindrance. Also, as the system develops, greater use of trains that do not stop at all stations and others that stop at only the main stations should further ensure that the line is used to its full potential (this already occurs to some degree on the East Coast line in Britain, but can be most clearly be seen with the use of Kodama, Hikari and Nozomi services on the Tokaido and Sanyo lines).

Undoubtedly one of the great advantages of developing such lines will be the possibility of implementing new and improved safety technologies. This will further reassure the public and encourage them to use the high-speed service. On top of this, the number of services should be significantly increased (at present many lines only have one train per hour). Although this may be loss-making in the short-run, the knowledge that another train will be available shortly if one is missed, is likely to further encourage train usage.

Although these lines does not allow for cross-country services, this will be less of a problem as the present network will be able to handle cross-country services more effectively (as suggested above) and the extra speed of using the new lines (although it involves the inconvenience of changing trains) will ensure that journey times are reduced on those routes also.

To put these suggestions into perspective. If we assume that the British high-speed services were to similar to the average a speed of the shinkansen (say around 150mph) then the journey times on these four lines would be as follows (journey times are in minutes with present journey times in brackets):

  • East Coast Line
  • Leicester 42 min. (70), Nottingham 59 (105), Sheffield 76 (135), Leeds 89 (170), Middlesbrough (118 not 227), Newcastle 132 (214), Edinburgh 173 (319), Aberdeen 222 (478)
  • West Coast Line
  • Birmingham 51 min. (114), Manchester 88 (190), Liverpool 102 (240), Glasgow 187 (296)
  • West & South Wales Line
  • Newbury 24 min. (47), Swindon 35 (59), Bath 49 (90), Bristol 54 (107), Cardiff 70 (126), Swansea 86 (192)
  • South Coast
  • Brighton 25 min. (49), Portsmouth 45 (85), Southampton 52 (72), Exeter 97 (161), Torquay 105 (222), Plymouth 118 (252)

In conclusion, the shinkansen is not the perfect system - but it certainly has many things to teach Britain. Unfortunately I have not been able to go into depth about all the areas of what makes the shinkansen work so well or what Britain and other countries could learn, but this paper has attempted to suggest some of the improvements that could restore confidence in the system. Although the areas where improvements can be made do not all require vast sums of money, some may - particularly in the building of new lines. However, developing a good railway network is like getting a good education, it may be expensive to do so, but not as expensive as not doing so ... it is time for Britain to bite the bullet!


This paper is based on a presentation given at the Royal Institute of International Affairs on 19 January, 2001. It was further updated and written up in April 2001 while travelling on the shinkansen between Osaka and Tokyo.


[1] The best example of this is perhaps the advertisement on British television that depicts the shinkansen going at 300mph - whereas in fact the model shown (300 series) only has an official top speed of 270kph.

[2] Byunbyun, accessed on 10 January 2001.

[3] I travelled on the old 0-series Dr Yellow only a few days before it was withdrawn from service in early 2001 and replaced with a new, improved (and brighter-yellow) 700-series Dr Yellow.

[4] Kondoh, K.,'Integrated Transport System', Japan Society Proceedings, Number 136, Winter 2000, page 25.

[5] Comment made during an interview in January 2001.

About the Author

Christopher P. Hood is Director of the Cardiff Japanese Studies Centre at Cardiff University, Associate Fellow of the Royal Institute of International Affairs, and Editorial Consultant to the electronic journal of contemporary japanese studies. Personal homepage.

Email the author

Back to top