WIREDGOV

During my masters at the University of Leeds, I lived very close to the buildings where my classes were held. I never used to think about public transport as I used to walk everywhere. But this all changed when I moved to another area for work. I checked transport links to my workplace before moving, but the day after I moved the train route was shut, owing to the Transpennine Route Upgrade. I had to use the bus to get to work, which was frustrating because of the frequent delays and cancellations.

From shopping trips to dropping off our children in school, when buying or renting a house, we all consider transport connectivity in making various decisions, even if it is only at an intuitive level. Research indicates that improving transport connectivity boosts economic growth, employee productivity and improves access to labour markets [1]. For individuals, better transport accessibility means access to more opportunities, social interactions, goods and services, and has also been linked to higher life satisfaction and wellbeing [2, 3].

I’ve been exploring the Commission’s connectivity dataset based on work done by Prospective, which developed connectivity metrics for around one thousand of the most populated places in Great Britain. An analysis of these metrics was carried out and reported in an initial discussion paper in 2019 [4]. Prospective then constructed a new dataset using updated census figuresadded additional granularity across transport modes and alternative spatial areas which was used as supporting evidence for the Second National Infrastructure Assessment. When I came across this dataset, the first thing I checked was the intra urban public transport connectivity of my hometown Leeds – I was not surprised to find it to be in the bottom decile of all cities in Great Britain.

The metrics

Connectivity

So, what exactly is connectivity? Transport connectivity is defined as “the extent to which passengers or freight flows from a node can reach other nodes directly (direct connection) or indirectly through another node or a series of nodes.”[2]. Essentially, it is a measure that gives us an idea as to how easily the movement of passengers and freight is facilitated between places through transport links. The dataset includes metrics for connectivity within places (intra urban), between places (inter urban) and between places and international gateways (like airports). For example, let’s consider travel within cities, intraurban travel. This was calculated using average travel times between all points in the built-up area (a suburb, for instance) and the centre of the area (defined as the centre of economic activity)[5] and was weighted by demand (population or employment).

Mathematically speaking, connectivity is a ratio between raw accessibility and crow fly accessibility (note 3). Raw accessibility is a function of observed speeds and the travel time between the economic centre of the region and other geographical units.  Since connectivity is a ratio, a value of one means both the raw and the crow fly accessibilities are the same (note 4).

Congestion

This metric is a ratio that compares accessibility between peak and off-peak times (note 5). Ideally, we want congestion to score 1, because it means there is no difference between travel time during peak and off-peak hours and therefore little-to-no observed congestion. Essentially, higher is better for this metric. A full breakdown is found in the Transport Connectivity document published by Prospective.[5]

Population and connectivity: What are the links?

If you want to move away from cities, but still want access to them by public transport, there are a handful of places you could move to. Maybe Berwick upon Tweed would be a good start (more on that later)? Why note explore our Tableau visualisations to find out how various places in the UK stack up against one another.

The Commission team carried out an exploratory analysis of the dataset for car and public transport connectivity in intra-urban and inter-urban settings (Figures 1, 2 & 3). These graphs only scratch the surface of what the dataset is capable of, and it was used by a researcher at the National Institute of Economics and Social Research for a policy paper. Indeed, we encourage other users to explore this dataset further.

FIGURE 1: INTRA AND INTER URBAN CAR CONNECTIVITY OF THE 973 MOST POPULATED PLACES IN GB (HIGHER IS BETTER)

Figure 1 shows the intra-urban and inter-urban connectivity metrics for cars in GB. Intra- urban peak connectivity is worse for cities. Both intra- urban peak and off-peak connectivity show a negative correlation with population. But we can see that the rest of the graphs show a more mixed relationship for cities and other places versus connectivity (note 2). Finally, we can also see that the plots for cities show a significant variation, compared to the relatively more tightly clustered plots for other places.

FIGURE 2: INTRA AND INTER URBAN CONGESTION (HIGHER IS BETTER)

Figure 2 shows congestion for intra-urban and inter-urban travel. For both intra- and inter-urban travel, population is negatively correlated with congestion. That is, higher population means worse congestion. Inter urban congestion displays greater variability across places of similar size, somewhat masking the overall negative correlation.

FIGURE 3: PUBLIC TRANSPORT CONNECTIVITY (INTRA-URBAN AND INTER-URBAN) (HIGHER IS BETTER)

Figure 3 shows the public transport connectivity in both intra-urban and inter-urban settings. As per Figure 1, the intra-urban connectivity metrics for public transport show that as population increases, the connectivity metric reduces. Cities fare worse in the intra- urban setting. Like before, the inter-urban connectivity has a greater degree of variation for places of similar size, but does not display a negative relationship between connectivity and population – in general, cities are better served than smaller places by long-distance road and rail networks, but there are clear exceptions. The small town of Berwick-upon-Tweed is the best performing place when it comes to inter-urban public transport connectivity. We can speculate that because it is on the border of England and Scotland, and is well connected by direct trains to Edinburgh, London, York, Leeds, and Newcastle (it lies on the East Coast Main Line) – it scored highly. A great example of a place for those who want to leave the city life behind but still want to have access to cities!

All roads lead to London

These graphs could be read alongside figures 4 and 5, which shows some key congestion and connectivity metrics for various places as an average.

FIGURE 4: SELECTED INTRA-URBAN CONNECTIVITY AND CONGESTION METRICS (HIGHER VALUES ARE BETTER)

FIGURE 5: SELECTED URBAN CONNECTIVITY AND CONGESTION METRICS (HIGHER VALUES ARE BETTER)

London shows poor connectivity across the board, except for public transport and inter-urban car off-peak. However, the Capital is the worst performing city in the UK when it comes to intra-urban peak congestion, likely related to slow road speeds. According to TomTom, the navigation equipment manufacturer, London is the most congested city in the world where the average time it takes to travel 10km is 36 minutes. This is significantly worse than Dublin, the second most congested city, where it takes around 28 minutes [6]. This could explain why London scored poorly on intra-urban metrics, since travel time is an important factor in the connectivity metric. The Globalisation and World Cities report says that when it comes to globalisation and interconnectivity, London is one of the most well-connected cities in the world, compared to the rest of the UK [7] (therefore, it is expected that London is well connected to cities in the UK by road and rail and performs well on inter urban metrics).

The off-peak car connectivity metric is greater than one for all places, on average. While a value higher than one might raise a few eyebrows, it should be no surprise that we can travel “faster than the crow flies”. In this case, it is because Prospective set the speed of crow-flies accessibility to 50km/h, while the statutory speed limit is 70 mph (~110km/h) – it may be further to take the motorway, but it is still faster overall. Finally, Figure 5 also show that public transport connectivity is poor across the country on average, compared to car connectivity, potentially signifying underinvestment in public transport infrastructure.

Final thoughts

This disparity between the connectivity of London and other cities at the inter-urban level was noted by the Commission in its Second National Infrastructure Assessment. It made a recommendation that the quality and reliability of transport infrastructure in other places must reach the level of London, to boost economic growth and living standards [8]. When it comes to connectivity and congestion, the UK still must work on a lot; in London, that would be to improve congestion; outside it, improving connectivity.

Kabith is a Graduate Analyst at the National Infrastructure Commission in Leeds, who joined from the Institute for Transport Studies at the University of Leeds. He now travels to work on a bicycle.

Notes

1. Places in this instance refers to Built-Up Areas, subdivisions and travel to work areas as statistically derived by the ONS [9], cities, as defined by the Centre for Cities [10], Local Authority Districts, regions, combined authorities, and counties.
2. A log scale for population was used to make comparisons more accurately between places with a population in the low thousands and large cities like London which exceeded 10 million. This had the effect of making less populated areas (which were higher in number) not get clustered in a single region of graphs.
3. connectivity = raw accessibility / crow fly accessibility
4. The values for inter urban and international gateways were calculated the same way. The only difference is that, for inter urban, the travel time was calculated between the centres of places (for instance, the centre of Glasgow to Leeds). For international gateways, it was between the centre of a place and the gateway.
5. congestion = car accessibility at peak time / car accessibility at off peak time.

References

1. Alstadt B, Weisbrod G, Cutler D, Relationship of Transportation Access and Connectivity to Local Economic Outcomes, Transportation Research Record: Journal of the Transportation Research Board (2012); 2297(1), pp 154–62.
2. Rodrigue, JP, (2024), The Geography of Transport Systems.
3. Chung S, White M, Abraham C, Skippon S, Commuting and wellbeing in London: The roles of commute mode and local public transport connectivity, Preventive Medicine (2016), pp 88; 182–8.
4. National Infrastructure Commission (2019), Transport Connectivity Discussion Paper
5. Prospective (2022), Transport Connectivity Methodology Report, cited in 2024
6. TomTom traffic index Ranking(2023)
7. Globalisation and World Cities, World Cities 2024, GaWC (2024)
8. ‌‌National Infrastructure Commission (2023), Second National Infrastructure Assessment
9. Office for National Statistics (2021), Census geography
10. Centre for Cities (2024), City by city.