Menu Close

How did Grouping affect staff safety? Railway Grouping & accidents pt 2

In our last post, we used our database of accidents to British and Irish railway staff to look at a few cases occurring immediately around the time of Grouping, in January 1923. This is very much one of the strengths of our project – we can get to the individual level and see the people involved, as well as what was happening on the ground, and sometimes what happened afterwards.

There’s also a power in going in the other direction – thinking about what all of the cases show us when we start to combine them. We can start to consider how far any one case might be typical or atypical. We might ask questions about which accidents happened time and time again – and why. We might be able to pick up patterns in terms of which locations, roles or companies were more or less subject to accidents than others.

In this blog post, we’re going to focus on 1923 and the railway Grouping. In what ways might the change in industry structure have had an impact on staff safety and accidents? What follows is very much preliminary thoughts, and rather superficial analysis – but it indicates the sorts of things our project database can allow us to consider. Ideally we’d compare pre- and post-Grouping details, but for the sake of time and space here we’ve focused only on the immediate post-Grouping period, for a few categories found in the database. It’s worth remembering, too, that we can only do this for those accidents which were investigated by the state inspectors – around 3% of all cases. How the inspectors selected which cases to investigate isn’t clear, so the data we’re left with is already to some degree subjective.

How were accidents distributed across the UK? It’s important to remember that at this point the UK consisted of England, Wales, Scotland and Northern Ireland. However, during 1923 no accidents in Northern Ireland were investigated by the inspectors. In addition, the situation there wasn’t affected in the same way by Grouping, something we’ll discuss in the next post in this series. For now, we can take two snapshots from the database – accidents investigated in January 1923 and in the whole of 1923.

Casualties investigated by nation, in January 1923 and for the whole of 1923.

Pie charts showing casualties in January 1923 and for the whole of 1923, for England, Wales & Scotland.

 

We can see a strong preponderance of accidents investigated in England – as was the case before Grouping. Amongst other factors, this reflects the greater density of railway coverage (including movement of stock) in England when compared to the other nations. It probably also reflects an England-centric approach from the inspectors. In this case it looks like we see relatively little impact from Grouping. As we might expect in the immediate post-Grouping era, the accidents carried on happening and didn’t suddenly change just because the overarching organisation of the industry altered. On the ground, there might have been new liveries and company names, but the operational practices remained much the same as before.

Pie charts for January 1923 and the whole of 1923, showing the distribution of accidents by railway company.
Distribution of accident investigations, by railway company, for January 1923 and all 1923.

 

How were casualties distributed across the new ‘Big Four’ companies? As the charts demonstrate, the accidents investigated in January 1923 reflected reasonably closely those investigated across the course of that first year of Grouping. (The ‘other’ category in both charts is something we’ll discuss in the next post in the series.) It appears that the larger companies – the LMS and LNER – came in for most attention, and the sole ‘pre-existing’ company, the GWR, might have received slightly less scrutiny than the new organisations. Was it trusted more than the new companies?

Bar chart, showing accident numbers by location/ activity.
Comparison of accidents investigated by location/ activity type, January 1923 and all 1923.

 

In terms of where the accidents happened/ the activity being undertaken in the first year of Grouping, we can see what we might expect to be a reasonably typical (for its time) distribution of accidents. Shunting remained the most dangerous activity a railway worker might be involved in. If we were to produce charts for the same data for other years, pre- and post-Grouping, they would likely look very similar in terms of percentages, though the numbers would be different (with more accidents the earlier back you go).

On one level this is all a rather superficial analysis. What we’re left with – and what are more difficult to answer – are a series of bigger questions. Most pressingly, in the longer term, what impact did combining the industry into fewer, bigger, companies have on staff safety and accidents?

Without much more detailed analysis, it’s impossible to say – and it might not be possible to do that analysis at this distance. Ideally it would correct for relative numbers of staff, route mileage and traffic type and density. Some of that information is very difficult, if not impossible, to obtain 100 years after the fact – if it were even collected at the time.

We can speculate that merging lots of smaller organisations into fewer but larger companies might have had benefits via economies of scale. That might have been an increased willingness to invest in staff safety or stronger safety cultures. Yet there might also have been an initial ‘shadow’ period, whilst the merger was put into practice on the ground and different ways of working potentially clashed and produced more accidents. Really detailed knowledge of pre-Grouping railway company operating practices might be compared with the post-Grouping accidents in the project database to see if there were any discernible accidents arising from differing ways of working.

To offer some initial comments on the types and density of traffic – not supported by detailed statistical information – the Big Four companies had rather different ‘characters.’ The SR was the smallest of the Grouped companies (in terms of staff and route mileage). A significant part of its traffic was focused on electrified suburban passenger services in London and surrounding areas, rather than heavy goods movement. These factors limited the risks staff were exposed to, and consequently the SR had the lowest portion of staff accidents in the Big Four. In comparison, the LMS and the LNER were far bigger, and served much more heavily industrialised areas. It’s therefore unsurprising that their staff accident records were – in absolute numbers, at least – rather worse than the SR and GWR.

Clearly this is a complicated topic, and needs rather more systematic analysis than we’re able to do at the moment. Nevertheless, we’ve started to use the project database to think about how the industry changed at Grouping, and the staff safety implications. In our final post in the series, we’ll look at some other things that didn’t change dramatically, or at all, in the Grouping – the situation in Northern Ireland, the continued existence of pre-Grouping companies in Britain, places like docks and organisations like light railways.

5 Comments

  1. Pingback:‘Never even blew me cap off!’: Railway Grouping & accidents pt 1 - Railway Work, Life & Death

  2. Peter Robinson

    Interesting the comparison of type of accidents showed that shunting was the most common which didn`t surprise me. While transcribing the data it seemed most of the accidents I was entering were shunting accidents or going between vehicles, also interesting that the workshops were where the least accidents happened, do you think that is because they were being overseen by the bosses. Have you done a comparison on the age of the staff that were injured or killed?

    • Mike Esbester

      Thanks Peter – and yes, no surprises about shunting, or anything relating to work that took staff near vehicles. In some cases – workshops will be a case in point, in a moment – the statistics from the reports give a misleading impression. However, with shunting they’re fairly representative of the risks of the work. As the reports only cover around 3% of all staff accidents for the period, the inspectors were selective about what they could investigate and therefore what we have today reflects that selection. We need to compare the overall (all accident) figures with those in the database to see how far what we’ve got in the database is an accurate reflection of the relative proportions of accidents by location/activity, vs how far it reflects particular interests of the inspectors.

      The demonstration of this is the workshop accidents. We’ve relatvely few in the database, not because workshop jobs were particularly safer, but because ‘our’ inspectors (the Railway Inspectorate) didn’t investigate accidents in workshops or railway factories! This was an odd jurisdictional thing, and workshops fell under the remit of the Factory Inspectorate, which was part of a different branch of the state (Home Office, rather than Board of Trade/Ministry of Transport). There’s more on this issue here: https://www.railwayaccidents.port.ac.uk/where-are-the-workshop-workers/

      That final question about age of staff injured/killed is a really good one – and at the moment, the answer is I’ve not looked at it! However, I’ve filed that one away to return to as it’s an interesting question: thank you! And of course, thanks as ever for all your work on the project – it’s much appreciated.

  3. Gordon Dudman

    Yes, understanding of age (and if possible correlation with length of service) would be interesting. I fear that the longer your service and age there is a dreaful tendency to take risks simply because you believe your experiences tell you its ‘OK’ to do that.

  4. Pingback:How didn’t Grouping affect mainline staff safety? Railway Grouping & accidents pt 3 - Railway Work, Life & Death

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.