Driving up Wages: The Effects of

Road Construction in Great Britain∗

Rosa Sanchis-Guarner† (LSE, SERC, Grantham)

February 2014

Abstract

Reductions in travel time between locations have direct consequences on the effect-
ive size of labour markets. However, there is little evidence on how road improvements
affect labour market performance. This paper estimates the effects of road construction
on individual outcomes using micro data from Great Britain. To capture the effects, I use
a measure of accessibility to employment through the road network. To address the en-
dogeneity of new roads placement, I compare workers located close to the schemes and
identify the effect from different exposure to accessibility changes. I further use work and
home location specific individual fixed-effects to separate the effects due to mobility from
those due to accessibility changes in-situ. For stable work-home locations, I find a positive
impact of increases in accessibility from work location on wages and hours worked, but
no effect of accessibility from home on either outcome.

Keywords: Job accessibility, Labour markets, Roads, Spatial sorting

JEL codes: J31, 018, R12

∗ The work in this paper is based on data from BSD, ASHE and NSPD produced by the Office for National Statistics
(ONS) and supplied by the Secure Data Service (SDS) at the UK Data Archive and by the Virtual Microdata Laboratory (VML).
The data is Crown copyright and reproduced with the permission of the controller of HMSO and Queen’s Printer for
Scotland. The use of the ONS statistical data in this work does not imply the endorsement of the ONS in relation to the
interpretation or analysis of the statistical data. This work uses research datasets which may not exactly reproduce National
Statistics aggregates. All the results have been granted final clearance by the staff of the VML and by the staff of the SDS.
Additionally, any interpretations or opinions expressed in this presentation are those of the authors and do not necessarily
reflect the views of the Department for Transport who also provided data to the study.
† I thank Henry Overman, Steve Gibbons, Laurent Gobillon, Nathaniel Baum-Snow, Diego Puga, Teemu Lyytikäinen,

Olmo Silva, Thierry Mayer, Guy Michaels, Gilles Duranton, Gabriel Ahlfeldt, Steve Ross, Ferdinand Rauch and Felix Wein-
hardt for comments and suggestions. I am also grateful to conference and seminar participants at the Urban Economics
Association Meeting 2011, the LSE Labour Markets Workshop 2012, the Finnish Government Institute for Economic Re-
search (VATT) 2012, the University of Valencia Seminar 2012, the Society of Labour Economics Meeting 2012, the 2nd IEB
Workshop in Urban Economics 2012, the 2nd European-UEA Meeting 2012, the IEB Seminar February 2013 and the CEPR
Conference in Regional and Urban Economics 2013 for comments. Richard Welpton and the staff at the SDS kindly helped
with the data. All errors and opinions are my own. Comments and suggestions are welcomed. I acknowledge the Bank of
Spain, the LSE and the ERSC grant ES/J007382/1 “Transport investments and spatial economic performance” for financial
support. Contact: Spatial Economics Research Centre, London School of Economics, Houghton Street, WC2A 2AE, Lon-
don, United Kingdom. Full draft and online appendix available at http://personal.lse.ac.uk/sanchisg/Rosas_
website/Research.html. Contact: m.r.sanchis-guarner-herrero@lse.ac.uk
1 Introduction
Road network is a hugely important part of infrastructure in all countries. In Great Britain roads
are the most important transportation mode both for goods and passengers. Over 90% of passen-
ger transport is done by road. In 2001 over 80% of the commutes in the UK were done using motor
vehicles, and this percentage increases in city size1 . New transport projects require large amounts
of public investment. Between 2001 and 2008, investment in road infrastructure increased in Great
Britain around 55% in nominal terms and around 40% in per capita terms. In 2008/09 over £16,000
million of public expenditure was invested in transport infrastructure, and around 40% of it on na-
tional and local roads2 . During this period, the major roads network was extended by around 175
kilometres.
Better transport infrastructure brings places and people closer together. This has two effects on the
actual size of markets. Firstly, for a given location of firms and workers, effective density increases,
as it becomes easier to reach other locations using the improved transportation network. Secondly,
new infrastructure changes the attractiveness of locations, which may boost spatial concentration
if firms and workers relocate. These effects may reinforce each other and create positive agglomer-
ation spillovers. Disentangling the empirical relationship between infrastructure construction and
economic outcomes is essential to the design of transport policy, and, given the importance of road
transportation for the movement of people and goods, the correct identification of the impact of road
investments is also important for economic policy as a whole. Yet, the number of empirical studies
that have tried to quantitatively assess these effects is still limited.
The objective of this paper is to provide causal estimates of the effects of transport improvements
on individual labour markets outcomes. I provide evidence for Great Britain between 2002–2008. To
measure the local impact of road construction, I combine data on road projects with micro data on
firms. For the outcome variables I use data on individual labour market outcomes for a panel of
workers. I test the effect that reductions in travel times induced by road construction have on their
wages and labour supply. My results provide the first causal evidence on the effects of changes in
proximity on individual labour market performance.
Building-up on my previous work in Gibbons et al. (2012), the effect of road construction is cap-
tured using an index of accessibility to employment (effective density). This measure is similar in
nature to “market-access” indices. It quantifies the amount of employment which is reachable using
the road network from a given location, inversely weighed by the travel time to reach other locations.
Most previous related studies rely on area variation to identify the effects of road infrastructure on
economic outcomes. They use measures such as density of roads within an area or area connectivity
to the network. These studies estimate the effects of road by comparing road endowment of different
locations over time3 . The accessibility index has important advantages over these measures. Firstly, it
is not constrained to artificial geographical boundaries. It captures how a location is affected by road
construction taking place over the whole geography. Secondly, accessibility captures how locations
are affected by changes in the road network even if these changes are small. Regardless of a location
being previously connected to the network, when new links reduce the travel time necessary to reach
other locations, they affect relative “connectivity”. The extent of the effect would depend on the eco-
nomic size of close-by locations and on the position of the area in the road network relative to the
rest of the country.
The geographical unit used as the basis for the calculation of the accessibility index is the elect-
oral ward. Wards (as defined in 1998) are very small areas and there are over 10,500 in Great Britain4 .
1 Exceptfor cities over with over 1 million inhabitants, where there is larger availability of public transport. Source: UK
Census 2001.
2 Source: Transport Statistics Great Britain and HM Treasury.
3 Some examples are Baum-Snow (2007a); Michaels (2008); Duranton & Turner (2011); Faber (2012). Notable exceptions

are Holl (2012) and Donaldson & Hornbeck (2013).

4 The average area of British wards is 21 square kilometres (they are smaller in England, 15 km2 , and larger in Scotland,

1
Having such a large number of units implies that there is a fair amount of variation in the values
of accessibility changes due to road construction and that they change almost continuously over
space. I exploit the fine geographical detail of the data to implement an empirical strategy capable
of identifying causal effects. I estimate the impact of accessibility changes on individual labour mar-
ket outcomes (mainly on wages and hours worked). I have information on workers home and job
locations over time, so in order to investigate the different channels through which accessibility can
impact individual labour market performance, I estimate both the effects of accessibility from home
and from work locations, conditional on predicted commuting travel time. To causally identify the
effects I follow an empirical strategy which is based on the use of individual micro datasets and small
geography.
The potential sources of bias are addressed in three steps. Firstly, cross-sectional estimates of the
effect of accessibility on labour market outcomes could be biased if the model does not capture un-
derlying time-invariant factors (such as individual specific productive advantages) that affect both
effective density and economic outcomes. I use a fixed-effects estimation method to address this
problem. I exploit the longitudinal nature of the data and control for individual heterogeneity. The
second identification challenge arises because over time accessibility in locations varies due to em-
ployment (economic size) and travel times (proximity) changes. If local employment changes as a
result of road construction5 , I would not be able to separate the effect of effective density changes
on labour markets arising from changes in local employment from that due to changes in proximity.
Furthermore, accessibility changes due to relocation of employment may be partly driven by the out-
come variable studied or be correlated with the same unobserved shocks. To address these concerns,
I construct the measure of accessibility fixing location employment to pre-period levels (2001), so it
only changes over time due to variations in travel times between locations. This way, I avoid using
spurious variation on effective density that arises due to endogenous changes in employment across
space in response to the new transport schemes. Additionally, by using this definition of accessibility,
I focus on the variation in accessibility over time stemming from changes in the channel of interest,
e.g. reductions in travel times between locations due to construction of new road links.
Finally, a major challenge to the identification of the effects is the fact that new road links are not
randomly allocated, but targeted specifically at locations to improve traffic flows, decrease congestion
or improve connectivity. For this reason, to identify the effect of roads on individual labour market
outcomes, we cannot simply compare workers located in areas that were “treated” with workers in
areas “non-treated”. The standard approach to address this issue is the use of instrumental variables,
normally based on historical routes placements or national infrastructure plans (used for example
by Baum-Snow, 2007a; Michaels, 2008; Duranton & Turner, 2011). Instead, I exploit the geographical
detail of the data and I only compare workers which are located close to the road schemes6 . When
new links are added to the network, optimal travel times decrease and employment accessibility
increases, but by different amounts according to where a place is in relation to the existing road
network, the characteristics of the new road links and the location of major centres of employment.
All the workers within this distance band are “treated” by the accessibility changes, but to different
extents and at different points in time, depending on their location. The coefficients are thus estimated
by comparing workers within certain distance from the road improvements. I identify the impact of
road construction by exploiting the treatment intensities, conditional on a large set controls. This
strategy is furthermore supported by the fact that the road links are aimed at larger areas than these
narrow distance band and aimed at connecting distant places and not at improving the local economy
(Highways Agency, 2009). It is quite unlikely that the new links are aimed at specific individuals in
wards within those narrowly defined distance bands, especially after controlling for different growth
trends around the schemes.
62 km2 ) and the average population estimate in 2001 (for England and Wales) is around 6,000 people.
5 A result that Gibbons et al. (2012) suggests.
6 This follows from Gibbons et al. (2012)

2
 By including individual fixed-effect, I estimate the coefficients using the within-individual changes
in accessibility. Over time, workers are exposed to different levels of employment accessibility de-
pending on their work and job locations. Within individual variation in accessibility could be due to
changes in accessibility for given locations (so stemming solely from changes in travel times) or due
to workers relocation over space (potentially as a result of new road links or due to reasons correlated
with them). As I follow workers work and home locations over time, different sets of fixed-effects (in-
dividual, individual-work, individual-home, individual-work-home) can be used. These control for
individual and area time-invariant unobservables and allow me to investigate the impact that work
and home mobility has on the estimates. The changes in the coefficient of work and home accessibil-
ity when I allow for different degrees of mobility informs about the role that spatial sorting plays on
identifying the impact of accessibility on individual labour market outcomes. When use I individual-
home-job fixed-effects (so I exploit the variation in accessibility for an individual while he is keeping
his work-home location pair fixed) I identify the effect of accessibility only from changes in travel
times for a given location. As mobility might be a result of road construction, this strategy isolates
the effect of changes in accessibility “in-situ” (pure effective density effects) from the effect due to
spatial sorting.
Allowing for spatial mobility, I find positive significant effects of accessibility from both work
and home on wages and hours worked, conditional on predicted commuting time. The effects could
be working via increased access in-situ or work or home relocation. When I use individual-work-
home fixed effects, I find positive significant effects of accessibility from work on earnings and hours
worked but I find no evidence that accessibility from home affects any of the outcomes. This suggests
that residential mobility is the source of variation from which I identify the impact of accessibility
from home on labour market outcomes. When I use the changes in accessibility for a given location,
only accessibility from work location has an effect on outcomes. Potential explanations of this finding
could be increased spatial competition for jobs (which results in higher individual performance) or
agglomeration externalities.
My research contributes to the existing evidence in three ways. First, I use very rich worker
microdata and a detailed dataset on road projects, which allows detailed study of the relationship
between transport improvements and individual labour market outcomes at a very small spatial
scale and to study several inter-related outcomes. To my knowledge, there exists almost no previous
evidence on the effect of road construction on individual labour market outcomes. Secondly, given
the quality of the data used, I can adopt a careful empirical strategy to tackle several identification
issues which could undermine the causal interpretation of the results. This improves the validity of
the results in order to draw policy recommendations. Finally, I provide robust empirical evidence
on some strands of the existing theoretical urban labour economics literature, as the paper provides
empirical evidence on the spatial mismatch theory and the effect of effective density on labour pro-
ductivity.
The remainder of the paper is structured as follows. Section 2 sets out the channels through which
transport investments can affect labour market outcomes and reviews the related theoretical and
empirical literature. The empirical strategy is explained in detail in Section 3. The data sources and
the construction of the variables used in the empirical analysis are also described in this section. The
results are fully explained in section 4, and robustness checks and an interpretation of the coefficients
is also provided. Finally, section 5 concludes and explains the future steps of this research. Tables and
figures from the online appendix are available online.

2 Theoretical framework and related literature

The role of transportation in the spatial distribution of economic activity and economic perform-
ance has become of increased interest to researchers in the last years. Decreasing transport costs are

3
considered to be a central driver of economic integration and of the rise of agglomeration externalit-
ies, but solid empirical evidence on the channels through which these effects operate is scarce. Even
if some authors have explicitly included the role of transportation into spatial economic analysis
(Combes & Lafourcade, 2001; Puga, 2002; Venables, 2007), there is still need to empirically establish
the causal link from transportation infrastructure to spatial economic performance, especially with
regard to the effects on individual outcomes.
While most of trade and location theory regards transport investments as having an effect on
(goods) transport costs (Michaels, 2008, for example), urban labour theories (Zenou, 2009) consider
transport policy as having an effect on commuting costs and therefore mostly operating through the
labour market. Glaeser & Kohlhase (2003) document the declining role of goods-transportation costs
in developed countries and highlight the increasingly important role of the mobility of workers in
modern economies.
Transport improvements affects labour markets through multiple channels (for a review see Gib-
bons & Machin, 2006). First of all, new transport links reduce commuting time costs. The “spatial
mismatch hypothesis” (SMH) predicts that larger physical distance7 between residential location and
job location has detrimental effects on the labour market outcomes of those living further away from
employment centres. A reduction in commuting time and costs associated with transport improve-
ments enables people to increase the scale of their job search and could also encourage potential
workers to participate in the labour market (Vickerman, 2002; Jiwattanakulpaisarn et al., 2009). For
unemployed or inactive workers reduced commuting costs decrease search costs and reservation
wages, and in this way, they can help mitigate frictional unemployment and increase the employ-
ment probabilities of those who are jobless (Inhanfeldt, 2006). By affecting optimal labour market
choices, shorter commutes also impact observed wages and unemployment rates (see Phillips, 2012,
for some evidence on these channels). If employers require workers with specific characteristics or
skills, employees can bargain to improve their labour market conditions (wages, hours worked, occu-
pation) to compensate for longer commutes. However, if labour markets are thin (Manning, 2006), i.e.
workers have access to a limited number of potential employers, longer commutes could not be fully
capitalised into nominal wages. Longer commutes can also have an effect on wages through their
effect on productivity, if shorter commutes are related to healthier or more motivated workers (van
Ommeren & Gutiérrez-i-Puigarnau, 2011). Evidence suggests that commuting costs have effects on
workers labour supply (Gutiérrez-i-Puigarnau & van Ommeren, 2010), absenteeism (van Ommeren
& Gutiérrez-i-Puigarnau, 2011) and wages (Manning, 2003; Mulalic et al., 2010).
Secondly, faster connections bring employers and workers closer together and hence affect the
effective size of the labour markets. Accessibility affects the tightness of the labour market, i.e. the
ratio of unemployed relative to the number of job vacancies (Détang-Dessendre & Gaigné, 2009). If
labour markets are more accessible to unemployed workers residing outside the labour market area,
or the labour market becomes effectively bigger because it is better connected, for a given number of
vacancies the number of potential candidates would increase. This increases competition for jobs and
might have two effects. On the one hand, unemployed workers living further away from job might
become employed due to the increase in accessibility8 . Manning & Petrongolo (2011) use British data
to study how “local” labour markets are and find that workers tend to look for and to accept jobs
which are relatively close to them, so labour markets are very local. Reduced commuting time due
to road construction can have an effect on the actual size of these local labour markets and affect the
number of competitors for a give search area. On the other hand, due to increased competition for
jobs and workers, we could see an increase in the quality of job matches which could be translated
into higher productivity and wages. Better matches can occur because for a given travel time workers
7 Distance includes geographical distance but could also include bad public transport provision or poor access to private

transport (car).
8 It is beyond the scope of this paper to study the effect of accessibility on employment. I use data for a panel of employ-

ees and this focus on labour market outcomes for employed workers.

4
have access to a larger pool of vacancies (Harmon, 2013).
Changes in effective density can also affect labour supply, e.g participation and hours worked. It
can weaken barriers to participation in the labour market, encouraging the entry of disadvantages
groups (female or low-skilled) into job search. It can affect hours worked if wages increase due to
transport improvements, making work more appealing; or even if increased competition in denser
areas induce young professionals to behave in a more rivalrous manner (the “rat race” argument in
Rosenthal & Strange, 2008).
Thirdly, reductions in travel times would increase the geographical scope of the agglomeration
economies, as for a given physical distance employers and employees are nearer to each other9 . If
agents are closer, there is more potential for interaction. Ahlfeldt & Wendland (2011), using the con-
struction of the railway in Berlin at the beginning of last century, show how distance becomes less
important and the scope of spatial interaction increases as transport infrastructure improves. The
emergence agglomeration externalities might have positive effects on workers earnings if urbanisa-
tion economies10 and proximity to markets (Krugman, 1991; Krugman & Venables, 1995) give raise
to productivity gains which are capitalised into nominal wages.
Finally, transport investments have an effect on location decisions. If better connectivity is per-
ceived as a locational amenity it would be taken into account when workers decide where to reside.
When transport investments are capitalised into housing prices (Gibbons & Machin, 2005; Ahlfeldt,
2011), they in turn affect real household incomes, and thus residential choices and labour market
behaviour.
In short, better transport connections can improve the efficiency of labour markets and have pos-
itive impacts on workers performance. Nevertheless, given the multiplicity of channels, the size and
direction of the effects of transport improvements on labour market outcomes remains mainly an
empirical question (Gibbons & Machin, 2006).
A number of recent papers have studied, with emphasis on the estimation of causal effects, the
impact of roads on a variety economic outcomes. Most of these papers use data for the US. Using the
1947 planned Interstate Highway System as an exogenous source of variation, Baum-Snow (2007b)
studies the effect of highways on the process of suburbanisation of American cities since the 50s,
and on the changes in commuting patterns since the 60s (Baum-Snow, 2010). Michaels (2008) uses a
similar source of exogenous variation to estimate the effect of reduced trade barriers on the demand
for skills. Hymel (2009) uses a cross section of US metropolitan areas to assess the impact of traffic
congestion on aggregate employment growth. Duranton & Turner (2011) examine the effect of road
construction on vehicle-kilometres traveled (VKT) in US cities. These same authors, using a similar
identification strategy, also estimate the effects of highway construction on urban growth (Duranton
& Turner, 2012) and on trade (Duranton et al., 2013). A few papers (Donaldson, 2013; Faber, 2012)
have focused on developing countries (railroads in colonial India and highways in China) to study
the effect of the reduction of transport costs, due to transport network development, on trade integ-
ration and resulting economic development.
With the aim of estimating causal impacts, my research is related to this recent strand of the lit-
erature. However, the identification strategy is different to the studies above. Most of these papers
use road network proxies at the metropolitan area level as their regressor of interest. In my case,
accessibility is computed at very small area level, so the treatment of space is quasi-continuous. Ad-
ditionally, the accessibility index changes smoothly over space even in areas where road construction
did not take place. Finally, my estimation methodology does not rely on the use of instrumental vari-
ables but exploits the geographical detail of the data by only comparing individuals affected by road
9 Rosenthal & Strange (2003, 2008) and Jofré-Monseny (2009) are some of the few studies which investigate the geo-
graphical scope of agglomeration economies. They however examine the geographical scope of the agglomeration effect,
not test the effect of agglomeration via changes in proximity.
10 Sharing, matching and learning in Duranton & Puga (2004) terminology; input-output linkages, labour market pooling

and knowledge spillovers in the classical Marshall (1890) terminology.

5
construction.
My findings add to existing evidence on the effects of agglomeration and market access on nom-
inal wages (for example Mion & Naticchioni, 2005; Combes et al., 2008). Most of these studies identify
the impact from local variations on employment density, but little attention has been given to the ef-
fect of agglomeration via changes in proximity between locations. New road links decrease travel
time between locations and, for given economic sizes, have a direct effect on effective density. In con-
trast with these papers, my paper provides evidence on how changes in “proximity” can give rise to
agglomeration externalities. If there are any potential productivity benefits to workers arising from
changes in accessibility, these could be induced through road constructions. Therefore, my estimates
inform not only on the overall effect of accessibility on worker economic outcomes but also on the
channel through which policy can potentially impact these outcomes. Additionally, my findings also
inform some of the predictions of the spatial mismatch literature (see Inhanfeldt, 2006, for a review)
and of the thin labour markets theory (Manning, 2003).
My main results explore the impact of road construction on workers pay, but I also examine the
effects on labour supply and investigate how spatial mobility affects the results. By looking at several
outcomes, this paper provided a wider analysis on the relationship between transport investments
and individual labour market outcomes than the existing evidence.

3 Research design
3.1 Measuring road construction
3.1.1 Accessibility measures

The aim of this paper is to estimate the causal effect of road construction on economic outcomes11 .
Therefore, the first challenge is to find a measure that captures changes in the road network. A variety
of measures have been used in the literature12 . As the current road network in Great Britain is very
dense and the length of the new links constructed is relatively short, using measures like changes in
connectivity or density of roads to capture the effects of road construction would not be appropriate.
Instead, I use a measure of accessibility to employment “through the road network” (or effective-
density) from each location at every point in time.
Accessibility to employment measures the amount of employment which is reachable using the
road network from a given location, inversely weighed by the travel time to reach these other loc-
ations. One advantage of using this measure is that it is not constrained to artificial geographical
boundaries like some of the alternative measures. Moreover, it allows us to use variation due to road
construction which affects optimal travel times, even if the location was previously connected to the
network. Additionally, it captures the effects of transport improvements over the whole geography.
This measure is, thus, appropriate in a setting where road density was already high at the beginning
of the period of study and where one aims to study the effect of additions to the existing network.
Formally, accessibility to employment Art from a given location r at time t is defined as:
R
∑ j 6 =r
 !

Art = a crjt ∗ econ size jt (1)

where a (.) is the transport cost function, crjt are the transport costs between locations r and j at time
t and econ size jt measures the economic size of the location at time t. Finally, t corresponds to years
11 The methodology described in this section builds from previous work I did with co-authors in Gibbons et al. (2012).

When necessary, the methodology was adapted to the specific context and research questions of the current paper.
12 Some measures are connectivity to the network (Faber, 2012), kilometres of roads within a given area (Melo et al., 2010;

Duranton et al., 2013), distance to closest highway (Baum-Snow, 2007a), number of rays crossing a given area (Baum-Snow,
2007a, 2010), presence of highways in a given location in a particular year (Chandra & Thompson, 2000; Michaels, 2008),
“lowest-cost route effective distance” (Donaldson, 2013) or amount of public expenditure on road infrastructure in a given
area (Fernald, 1999).

6
2002 to 200813 . I use electoral wards (as defined in 1998) as the geographical unit.
This index is a measure of the economic mass accessible to a firm or a worker in a particular loc-
ation, given the local transport network. At a given origin location r at time t, accessibility Art is a
weighted sum of economic mass in all destinations j that can be reached from origin r by incurring
a transport cost crjt along some specified route between r and j (for example straight line distance or
minimum cost route along a transport network – measured in travel time or in distance). The func-
tion a (.) determines how the weights enter in the calculation of Art . The economic size of locations is
measured using total ward employment and travel time between wards is used as measure of trans-
port costs (further details on the construction of ward-to-ward travel times and ward employment is
provided in Section A.1.). Assuming inverse cost weights14 and a cost decay equal to one15 , the final
expression for accessibility to employment becomes:
R
∑ j 6 =r
!

Art = 1/travel timerjt ∗ employment jt (2)

The accessibility index as defined in (2) is similar in structure to market potential measures used
in economic geography (e.g. Harris, 1954; Krugman, 1991), and to the accessibility indices used more
generally in the transport literature (e.g. Vickerman et al., 1999; El-Geneidy & Levinson, 2006). The
calculation of this index requires the construction of an origin–destination (O–D) matrix whose com-
ponents are travel times between the locations. When computing the O–D matrix I apply a limit of 75
minutes drive time (1.25 hours). This limit facilitates O–D matrix computation but hardly affects the
value of the accessibility index because wards beyond 75 minutes have negligible weights in the cal-
culation of Art . Moreover, as shown in Table A.1, more than 99% of commutes in the Great Britain are
below 90 minutes. I also exclude location r from the calculation of the accessibility measures to lessen
the potential reverse causality problem. This problem arises because the dependent variables include
labour market outcomes the individuals located in the ward and it is likely that these outcomes and
the economic size of a given location are be jointly determined. As explained below, I use an altern-
ative definition of accessibility to address this issue, so the inclusion of the own economic-size in the
calculation of Art would not be a major issue.
Accessibility Art changes in a given origin r are driven both by changes in travel times between
wards (stemming from road construction) and by changes in the employment of origin ward r and
in the different wards j around r. This may lead to endogeneity problems in the estimation of the
effect of accessibility, if the employment changes near the origin are causally linked with changes in
the economic outcomes in the origin or driven by the same unobserved factors. Moreover, as I focus
on the effect of road construction, the examination of Art would be limited as it would be impossible
to differentiate between the changes driven by employment relocation and the changes driven by
changes in travel time between locations. It is thus useful to construct an alternative accessibility
measure Ârt that focuses on the changes in accessibility that stem only from road construction:
R
∑ j 6 =r
!

Ârt = 1/travel timerjt ∗ employment jt0 (3)
13 As explained in Section A.1.2., road networks and travel times are dated at the beginning of the year. As the ward
employment data is dated in April of each year, in practice I combine travel time data at the beginning of year t (for
example 2002) with employment data in April of the previous year t − 1 (April 2001). This implies that in the empirical
specifications of section 3.2, accessibility measures are in fact lagged one year with respect to the outcome variables.
14 In the definition of A the value of the weight a (.) attached to any destination r is a decreasing function of the cost of
rt
reaching destination
  j from origin r. α is the cost decay. Potential weighting schemes include: “cumulative
  opportunities”
 
weights a crjt = 1 if j is within a specified distance of r, zero otherwise; “exponential weights” a crjt = exp αcrjt ;
     
−α
“logistic weights” a crjt = [1 + exp −αcrjt ]−1 or “inverse cost weights” a crjt = crjt . See Graham et al. (2009) for
further discussion of these indices.
15 Graham et al. (2009), using the inverse cost weighting scheme, estimate the parameter of distance decay functions for

several sectors using similar British data to ours and find values between 1.8 and 1, depending on the sector. The estimated
alpha for the whole economy is around 1.6. In practice, I check the robustness of the results to different distance decays.

7
where t0 is some fixed period of time at the beginning of the period of analysis. In the empirical
analysis below t0 corresponds to 2001. Fixing employment to its t0 level ensures that changes in the
accessibility index (3) over time occur only as a result of changes in the costs crjt (e.g. travel time)
and not as a results of changes in wards employment. By default, changes in Ârt are always positive
as employment is fixed over time and changes in the denominator are always reductions in optimal
travel times (see more details on the construction of these in the online appendix). In the empirical
work, I use Ârt in order to use only the variation in accessibility which stems from the road con-
struction. In the robustness checks I also instrument Art with Ârt , and the results remain unchanged.
Section A.1. in the online appendix contains full details on the construction of the ward employment
and the travel time between locations used in the calculation of the accessibility measures.
The empirical methodology uses the changes in the accessibility index at each location r to es-
timate the extent to which individuals in location r are “potentially” affected by the construction of
new roads close to them. The construction of Ârt is based on predicted optimal travel times which
change due to new routing after the construction of new links in the road network. I do not use in-
formation on how individuals (works and firms) actually use roads and which are the actual travel
times between locations. In this sense, the measure (changes in accessibility) captures the “intention-
to-treat” of road construction. Nevertheless, it is road construction (and not actual observed travel
times) that is the instrument used by transport policy to improve locations connectivity and alleviate
traffic congestion. In this sense, using changes in predicted travel times is the appropriate measure
to evaluate the effect to this transport policy.
Over all possible improvements through which transport investments can affect road networks
I focus on construction for two reasons. First of all, among all the different type of interventions
(e.g. dualling, resurfacing, improving lighting and signalling, etc), construction of new links is a
clear source of variation which is easily identifiable and measurable. Construction of new road links
affects optimal routes along the road networks and it constitutes the source which is likely to have
the larger impact on the variation in travel times between locations. It also is a well-defined tool
which can be used by policy to influence proximity between locations. The second reason is data
availability. The collection of reliable data on transport improvements and projects over a length of
time is demanding and their measurement can be problematic. The methodology I use to generate
variation in optimal travel times needs a clear definition of changes in the road network and addition
new links to the network provides a reasonably source of variation (see Sections A.1.2., A.1.3. and
A.1.4. in the online appendix for more information). Nevertheless, one must bear in mind that other
type transport improvements might also affect optimal travel times between location and the analysis
is limited from that perspective.

3.1.2 Descriptive statistics

Table 1 provides summary statistics on wards’ log accessibility. The top panel displays the statistics
for the change log accessibility between 2002 and 2008 (growth rate)16 . The bottom panel displays
values for the statistics for the annual log accessibility, which is the variation I use in the empirical
estimations. The table shows statistics for the accessibility indices for all the wards in Great Britain
(10,540) and for wards which are situated within 5, 10, 20 and 30 kilometres of the road schemes
carried out during the period of analysis17 . In the main estimations I use observations from workers
located in wards within 30 kilometres of the new links, but here I show summary statistics for smaller
distance bands in order to illustrate how accessibility changes increase the closer we get to the new
links. The accessibility measure is calculated fixing employment at 2001 levels ( Ârt ), so its variation
16 I calculate the growth rate between the first and last period of the analysis for descriptive purposes, but I do not use it

in the empirical analysis.

17 Along the road scheme. When schemes are close to each other, these bands overlap and are combined into a single

area. See figure 1(c) for more detail.

8
stems from changes in predicted travel times (denominator) and not from changes in ward employ-
ment. For this reason its range of variation is much lower than standard market-access measures.

[INSERT TABLE 1 HERE]

The upper panel of Table 1 shows that, for the whole Great Britain, between the first and last year
of the period of analysis employment accessibility induced by road construction was on average
small, only 0.3% (which is very similar to the decrease in travel times observed in Table A.12). How-
ever, average accessibility change increases significantly when we focus on wards closer to projects.
For example, within the 5-kilometre distance band the mean change is 3.5% and the 90th percentile is
8.5%. As we expand the sample away from the schemes changes in accessibility tend to fall. Within
10 kilometres, mean accessibility change is around 2% and the 90th percentile is almost 5%. Within
20 kilometres these values decrease to around 1% and 2%, and within 30 kilometres of the schemes
mean accessibility change is 0.75% and the 90th percentile is 1.5%. It is worth noting that the standard
variation of the changes is relatively large, suggesting a substantial amount of spatial variation of
changes in accessibility over space.
In the estimation of the empirical results, I exploit the annual variation in road accessibility. For
this reason, the bottom panel of Table 1 provides summary statistics on the level of log accessibility
in every year. Focusing on the statistics for the 30-kilometre band we see that the even if the overall
standard deviation is not very large, there is a quite a large range of variation between the minimum
value and the maximum value of log accessibility. In fact, in the panel estimates of section 4, I exploit
the within-variation of the variable. For this reason, the between and within standard variation of log
accessibility is also provided in the table. We notice that most of the overall variation stems from the
differences across wards, and this increases the further away we move from the schemes. However, as
we see in the empirical results, the amount of within variation of the variable within the 30-kilometre
band is sufficient to yield precise estimates.
Figures 1(a) and 1(b) illustrate the spatial relationship between the location of road schemes and
resulting accessibility increases. Figure 1(b) shows the changes in log accessibility between 2002 and
2008 which stem only from road construction ( Ârt ). The biggest changes in accessibility are around
the schemes plotted in Figure 1(a), but there is substantial spatial variation across the country. The
amount of spatial variation is more evident in Figure 1(c). It represents the same values as Figure
1(b) but focuses on the Manchester-Bradford-Leeds area. This figure shows the small scale of the
spatial data and also the substantial amount of variation in accessibility values close to the new
links. It also allows us to illustrate the identification strategy (which is explained in Section 3.2). The
location of the new links constructed between 2002 and 2007 are indicated by bold white lines, with
the name of the road labeled. The dark grey lines are ward boundaries. The thick black lines delimits
the wards which are included within 30 kilometres of the road projects within the area. The map
illustrates that the effects of road construction on accessibility vary considerably across wards in the
vicinity of the same scheme. In the identification strategy I argue that these differences in accessibility
changes across wards are coincidental and can be treated as exogenous, especially when controlling
for differential time trends near different schemes. Figure A.1 shows the geographical extent of these
distance bands.

[INSERT FIGURES 1(a), 1(b) and 1(c) HERE]

3.2 Empirical specification and identification strategy

The Annual Survey of Hours and Earnings (ASHE) is used to obtain results on individual earnings
and hours worked (more details in Section 4). I use a panel of workers to estimate the effect of access-
ibility to employment, both from work and from home, on individual labour market outcomes. The
main regressions use data for years 2002 to 2008. In the main results I focus on the effect of accessib-

9
ility on wages, as most of the literature on the effect of effective density on labour market outcomes
has focused on this variable. In section 4.3.1 I provide evidence on alternative outcomes in order
to investigate the multiple channels through which road construction might affect labour markets.
Without loss of generality, the notation below is refers to wages
The estimated relationship is:

yihwt = β 1 Aiht + β 2 Aiwt + β 3 cihwt + θXit + δZht + λWwt + µi + ξ t + ε ihwt (4)

The wage of a worker i living in ward h and working in ward w at time t is denoted with yihwt . Given
that accessibility is measured at the ward level, I initially ignore changes of home or job within the
wards18 . At each point in time, workers work and live in specific wards. Over time, workers location
can change, if they change jobs, change home or change both. Initially, I use individual variation
in accessibility and wages which could also be driven by spatial relocations (more on this below).
Aiht denotes accessibility to employment from home ward h at time t, Aiwt denotes accessibility to
employment from work ward w at time t19 , cihwt denotes the commuting costs (travel time) between
work and home at time t, Xit is a vector of personal and job characteristics, Zht is a matrix of home
ward characteristics and Wwt is a matrix of work ward characteristics. ξ t are year-industry dummies
that control for year-industry specific shocks affecting all wards in a given year. ε ihwt is the idiosyn-
cratic error. Both the labour market outcomes and the accessibility indices are transformed to natural
logarithms so we can interpret their coefficients as elasticities. Accessibility and commuting time are
lagged with respect to the outcome variable, as labour market outcomes measured in April of year
t, travel times are measured in January of year t and the economic size of locations is fixed to April
2001 employment.
The ASHE dataset does not report information on travel time or distance travelled to workplace,
but provides detailed information on the work and home location at every point in time. I use the
road networks between 2002 and 2008 (created as explained in the online appendix) to calculate
optimal travel time between ward of home and ward of work along the road network. We do not
have any information on the travel mode of the workers. It could be the case that they are not com-
muting by road or not commuting using the optimal route predicted by the GIS software. However,
using optimal travel time to proxy for commuting costs has the advantage of getting rid of poten-
tial measurement error on self-reported travel time20 . In practice, I estimate a reduced-form effect of
commuting costs, in which the measure of commuting cost is the optimal travel time through the
road network. Moreover, as this travel time measure changes over time due to the road construc-
tion, it does not drop out when I include individual-home ward-work ward fixed-effects below, as
opposed to straight geodesic distance. I include this information in the estimated specification (4) in
order to capture the effect of commuting costs on labour market outcomes. This way I can estim-
ate the effect of accessibility from work and home conditional on commuting costs. This helps to
interpret the results given the numerous theoretical channels through which transport construction
can affect labour market outcomes, as discussed above. For example, some of the effects of transport
policy on wages could be due to employers compensating workers for longer commutes and some
could come through increased spatial competition or agglomeration externalities. By controlling for
18 For feasibility, accessibility is calculated using a very large but limited number of geographical units (over 10,000). This

geographical division is a good approximation of the continuity of space as most home and job relocations induces a ward
relocation. However, workers can move houses or jobs within a given ward. As I have information on the home postcode
and the work plant (as a combination of firm-sector-postcode), in section 4.3 I also check the robustness of the result to
allowing or restricting home and plant movements within wards.
19 For generality, I define accessibility as in expression (2), therefore allowing for changes in both ward employment and

in travel times over time. As explained below, in the formulation of accessibility used in the empirical application I keep
ward employment fixed to 2001 values, focusing on the variation stemming only from changes in predicted travel times.
20 Which is common, as noted for example by Gutiérrez-i-Puigarnau & van Ommeren (2010). Indeed, when if we tab-

ulate the answers on commute travel times using data from the UK Labour Force Survey –non reported–, responses are
disproportionably cumulated in values such 5, 10, 15, 20 and 30 minutes, likely due to rounding-up.

10
commuting costs in equation (4) we can be more certain that the effects of accessibility are not due
to compensation for longer commutes as this explicitly controlled for. Moreover, the interpretation
of the coefficient β 3 also informs us about the relationship between varying commuting costs and
labour market outcomes.
I am interested in the consistent estimation of parameters β 1 and β 2 . There might be (time invari-
ant) unobservable individual characteristics that affect both individual labour market outcomes and
accessibility indices (in levels) at the same time, and that are not included in Xit . For example, more
able individuals may live or work in areas where accessibility and wages are higher. I include worker
fixed effects µi to control for this (which in practice is equivalent to estimate the demeaned model).
By introducing the individual fixed effects I estimate coefficients β 1 and β 2 using the variation of ac-
cessibility over time with respect to the average individual accessibility in the period of observation.
At each point in time the worker may hold different jobs in different work locations (w) and live in
different places (h). Therefore the level of employment accessibility to which the worker is exposed at
home and at work varies over time for three reasons: when work-home location changes, when ward
employment changes and when road construction takes place (changing predicted travel times).
If workers are sorting spatially in order to take advantage of the changes in accessibility we would
not be able to identify the separate effects on labour market outcomes which stem from spatial sorting
and those which are due to changes in accessibility for a given location (for example agglomeration
externalities). Sorting could be an outcome of accessibility or could be due to other unobservable
reasons correlated with individual labour market outcomes. For example, if workers with higher
ability move to areas where accessibility and wages are growing faster, then the correlation between
the changes in accessibility and the (demeaned) error term could be different from zero. The same
could occur if more able workers choose jobs in areas where wages and accessibility are growing pro-
cyclically. For these reasons, even after controlling for unobservable time invariant characteristics of
the individuals, there would still be reasons to think the estimates of β 1 and β 2 could be biased.
To investigate this issue I define individual home-ward-work-ward specific fixed effects, µihw and
exploit the individual variation in accessibility for a given work-home (ward) location. This strategy
is useful for two reasons. Firstly, it isolates the effect of accessibility on labour markets for a “given-
location”. When I use within-individual variation, I am identifying the effects of accessibility as a
combination of changes due to spatial relocations and changes in effective density in-situ. Apart
from the endogeneity concerns this type of variation raises (if movements are correlated with labour
market performance and changes in effective density), most of the theoretical predictions of the effect
of accessibility on labour markets apply to individuals who do not move over space, and thus this
strategy allows me to link the results with the theoretical explanations in a more direct way. Secondly,
comparing the estimates when I allow for different degrees of spatial mobility informs about the role
that spatial mobility plays on identifying the impact of accessibility on individual labour market
outcomes. Finally, individual-location specific fixed-effect additionally control for location-specific
unobservables which might be correlated with labour market performance and changes in predicted
travel time, for example unobservable ward characteristics which might have determined the loca-
tion of new road links on those specific locations.
Replacing the individual fixed effects in (4) with µihw gives:

yihwt = β 1 Aiht + β 2 Aiwt + β 3 cihwt + θXit + δZht + λWwt + µihw + ξ t + ε ihwt (5)

Rewriting equation (5) in demeaned terms by subtracting the individual-fixed location means across
time (focusing only on the accessibility measures):

(yihwt − ω̄i. ) = β 1 ( Aiht − Āi. ) + β 2 ( Aiwt − Āi. ) + (ε ihwt − ε̄ i. ) (6)

The effects of accessibility on individual labour market outcomes in (5) are obtained by exploiting

11
the changes of accessibility over time for an individual while keeping their work and home locations
constant. When including µihw , coefficients β 1 and β 2 are not identify out of variation stemming from
spatial relocations. This reduces the endogeneity bias that might arise if the relocations are endo-
genously determined after the changes in accessibility take place (that could be caused by increased
accessibility or other unobserved reasons). The comparison of the estimates obtained by estimating
(4) and (5) inform us about how residential and work location sorting affects the estimation of the
accessibility effects.
To implement this, each individual in the panel is allocated a different individual-location fixed-
effect depending on where he works and where he lives. The same individual can get several fixed
effects depending on his work-home location. In the data, we can identify two types of individuals:
those that never move work-home location pair (while observed in the data) and those that eventu-
ally change ward of work, ward of residence or both. Individuals in this last group may be spatially
relocating (sorting) due to changes in accessibility, so in contrast to (4), specification (5) does not use
this variation for the estimation of the effects of accessibility but includes both types of individuals in
the estimation sample. Using individual-location specific fixed effects is different in nature from re-
stricting the sample to individuals that do not move locations over time, as in the second case we are
selecting our sample (and therefore might introduce sample selection bias) while in the first case we
use the whole sample but exploit variations over time around mean accessibility values for individu-
als while their keep the same work-home locations. By not restricting the sample to non-movers, for
the same individual I exploit both the within variation for a given location and the between variation
for different individual-location observations, so for the same worker I am able to exploit a larger
range of variation to identify the effects. Restricting the sample to non-movers drops the information
for individuals that move and thus the estimation of the effects relies only in a sub-sample of workers
which might rise precision and sample selection problems21 .
However, there are still other possible sources of bias in the estimation of the effect of accessib-
ility using (5). Even when keeping the work-home location fixed, accessibility changes around the
individual means due to changes in employment (numerator) and in travel times due to road con-
struction (denominator). If changes in labour market outcomes, for example wages, affect accessib-
ility changes by means of attracting workers to wards around the ones in which the worker lives or
works, a reverse causality problem could challenge the validity of the estimates. There could also
exist unobservable trends which affect both the location of employment and the labour market out-
comes which could bias the estimates. As explained in Section 3.1.1, to overcome this issue, I calculate
Ârt , accessibility in t, as in equation (3), where ward employment is fixed to 2001 values and travel
times correspond to predicted road optimal values in January of t. Over time, the variation in this
measure only occurs when new road links are constructed and predicted travel times between wards
change. In practice, this is similar to instrumenting accessibility Art with Ârt , but using a reduced-
form approach. Using the reduced form approach allows me to focus on the variation in access which
I want to study (that stemming from road construction) and, as results in Section 4 show, does not
provides very different estimates from a two-step instrumental variables estimation.
Possibly the most important threat for the causal interpretation of the estimates is the fact that
new links are not randomly allocated to locations, but specifically aimed at certain places in order to
affect their economic performance. It could be argued that the approach is not valid if the construc-
tion of new links is aimed at areas which are experiencing unobservable shock which are correlated
with individual labour market outcomes. Transport investments may be taking place in areas in
which workers would have done better anyway. Most of the literature on the evaluation of transport
projects has devoted a large discussion about how to tackle the endogenous placement of transport
21 Forrobustness, in section 4.2 I obtain estimates of equation (4) restricting the sample to individuals that do not move
locations over time and, even if the results are qualitatively similar to restricting the sample to non-movers, I avoid intro-
ducing a potential sample selection bias in the estimates. Table A.3 in the online appendix report some summary statistics
(mean, standard deviation) for the movers and non-movers samples for initial values of some variables.

12
investment and most authors have dealt with this problem using instrumental variables estimation
(Baum-Snow, 2007a; Michaels, 2008; Duranton & Turner, 2011, are good examples.). Instead, I rely in
two things in order to tackle this identification issue. Firstly, according to the Department for Trans-
port documentation about the schemes, improving very local economic outcomes is not one of the
key objectives of the road projects carried out by the Government22 . Transport projects are generally
aimed at a large spatial scales and designed to improve safety or reduce congestion within a wider
area23 . As projects are aimed at connecting areas (for reasons related or not to economic perform-
ance), I exclude observations in wards situated at the beginning and at the end of the road construc-
tion schemes as these locations might have been specifically aimed by transport policy (similarly to
Faber, 2012).
However, even in these non targeted areas, it could still be the case that some individuals “treated”
by new road links and others are not because of unobserved reasons. One cannot simply compare
individuals which are treated to those who are not. I rely on the small geography and micro data
nature of the data and only compare workers who are close to the new road links and that experi-
enced changes in accessibility to different extents and at different points in time, depending on the
size of the closest road scheme, its location within the network and when it was completed. I define
a 30-kilometre wide distance band24 around the 23 schemes undertaken during the study period and
only compare individuals located in wards within this band. I identify the impact of road construc-
tion by exploiting the treatment intensities within this distance band, conditional on controls and
individual-location fixed effects. Individuals placed within a given distance band are more likely to
be exposed to similar shocks and road projects are quite unlikely to be aimed at specific individu-
als within this narrowly defined distance band. This strategy implies that changes in accessibility
(predicted travel times), conditional on controls, can be considered exogenous to individual workers
located close to the new links. Moreover, even if the placement of the new roads would be endogen-
ous to the individual labour market outcomes, its exact placing within the road network, and thus
how the accessibility of the wards is affected by the new link, is likely to be exogenous to individual
workers. Figure A.1 in the online appendix shows the extent and location of the wards within this
band. Distance bands are defined for both work and home location, but I experiment with the over-
lap of these bands in the robustness section (4.3) by restricting the commute distance between work
and home.
In order to ensure that the accessibility index Ârt is uncorrelated with the underlying area trends,
I further control for differential trends in the vicinity of road schemes. I control for the initial level of
accessibility interacted with a linear trend. I include a set of nearest scheme dummies (23 schemes)
interacted with year in equation (5) and I also control for the distance to the closest scheme within
the distance bands and a dummy which indicates if the scheme has been opened in the year of ob-
servation. These three sets of controls (closest scheme trend, distance to closest scheme trends and
opened scheme trend) capture any unobservable trends specific to the scheme (shape, length, open-
ing date) which could be correlated with the placement of the scheme and initial employment values
and future labour market outcomes around the vicinity of the road project. I also control for dif-
ferential trends of the wards (work and home) based on 2001 characteristics. I used CENSUS data
(provided by CASWEB) at the ward level to calculate the share of population aged 15-64 (workforce)
with higher education, mean age of population, share of population living on social housing, the rate
of unemployment, proportion of workers commuting using motor vehicles and the average distance
traveled to work. I also calculated a residential density measure, using address counts data in 2001
from the National Statistics Postcode Directory (NSPD) and the area of the wards in square kilo-
22 More details on British Transport Policy context are provided in Sanchis-Guarner (2012).
23 Forexample, the objectives of scheme “A6 Clapham Bypass” in Table A.13 were to “improve road safety”, “relieve
congestion” and “provide the opportunity for environmental improvement in Clapham by removing through traffic”. See
http://www.highways.gov.uk/roads/projects/6006.aspx for more information and the evaluation report.
24 Narrower bands are used in the robustness checks.

13
metres, obtained from EDINA-UKBORDERS. Finally, I control for “travel-behaviour” in these wards
controlling for the proportion of commuters that use motor vehicles and the average commuting dis-
tance. As they are time invariant, I interact the 2001 characteristics with a linear trend. These trends
control for differential growth in labour market outcomes, e.g. wages, depending on the level of these
ward characteristics in 2001, before the period of analysis which starts in 2002. All the specifications
include year-industry fixed effects to control for year industry-specific shocks. In the estimation of
(4) I introduce these ward characteristics in levels.
I furthermore control for individual personal and job characteristics. Some of these characteristics,
for example full time status or occupation, could be regarded as “bad controls” because they could
be outcomes of the transport policy. To help address this issue, I define the level of the characteristics
at the beginning of the period (the first time the individual is observed within each of the two panel
definitions) and I interact that level with a time-trend. By doing this I control for differential trends
in the evolution of the labour outcome depending on the initial level of the job and personal char-
acteristics. I use occupation (9 categories defined below), age group (10 year groups, from 16 to 65),
full-time status, plant-type status (private sector, propietor/partner, public sector, non-profit sector),
firm regulated by collective agreement dummy and gender trends; and firm size (number of em-
ployees in the firm). Finally, I also control by commute distance dummies (5 kms bands) interacted
with a linear trend. These control for different trends on individual labour market outcomes around
the schemes depending on how long apart are work and home locations. Without loss of generality,
to define the commute distance dummies I use straight-line distance between work and home post-
codes. In the estimation of specification (4) I introduce these individual characteristics in levels, and
in the estimation of (5), I introduce them interacted with a linear trend (except for firm size).
Table 2 reports some balancing tests to check the validity of the empirical approach. I only report
the tests for work ward, but the results for home ward are very similar. I test if a measure of economic
performance of a given ward at the beginning of the period of analysis (2001 or 2002 depending on
the availability of the data) is systematically correlated with the distance to the closest road scheme,
within the 30-kilometre distance band (excluding the wards at the beginning and the end of the
schemes). I include 2001 ward census controls and scheme dummies, and I cluster the standard errors
at the ward level. I test the effect of distance within the band on initial ward accessibility, average
weekly wages, average hourly earnings, average total hours worked, average commute distance (as
crow-fly) and average commute travel time (as predicted by the network given work and home ward
locations). The coefficients for distance to closest scheme are insignificant in all cases (only weakly
significant for commute travel time) and close to zero. This indicates that, conditional on scheme
dummies and census controls, there is not systematic relationship between how close the wards is to
a given scheme within the distance band, so the variation of accessibility within the distance band is
“as-good-as-random”.

[INSERT TABLE 2 HERE]

Finally, in the estimation of the empirical results I cluster the standard errors to correct for arbit-
rary within-group correlation of the individual shocks in two distinct non-nested categories. In the
case of the estimation of specification (4), I cluster at the individual level. In the case of the estimation
of (5), I implement 2-way clustering at the home-ward and work-ward level. By doing this, I allow
the errors of the workers to be correlated at the treatment level (wards). In addition, the standard
errors are also robust to arbitrary heteroskedasticity.
There are a few limitations to this approach. The network data is simplified due to data limita-
tions, and some assumptions on the impact of new links additions on travel times are made in order
to be able to create a clear source of variation on travel times (full details are provided in the online
appendix). I focus on additions of new links to the network and disregard other road improvements
like dualling and resurfacing, which might also have positive effect on minimum travel times. Fi-

14
nally, the level of traffic congestion is fixed in the construction of predicted travel times to be able to
separate the policy instrument (more road kilometres) from the policy objective (reduce congestion).
Given these limitations, there is likely attenuation bias due to measurement error in the calculation
of the accessibility index, so my estimates would be a lower-bound of the real impacts.
One limitation of the paper is the lack of substantial evidence on how accessibility directly af-
fects work and home spatial sorting. Using individual-work-home fixed-effects exploits the variation
stemming only from changes in accessibility for a given work-home location pair choice. Compar-
ing the results using this approach and a standard individual fixed-effects estimation (which allows
variation in accessibility to arise from job and home relocations) sheds some light on the effect that
spatial mobility has on the source of variation of accessibility when identifying the coefficients. But
the causal relationship between accessibility and work and home location is relevant on its own and
requires further investigation.
Finally, my approach does not take into account that there might be other externalities stemming
from the road construction, for example changes in congestion or pollution. Evidence (Duranton &
Turner, 2011) suggests the elasticity of traffic flows with respect to new highways is close to one, so
it is quite likely that congestion was substantially relieved after the road construction. Nevertheless
more evidence on the impact of road construction on traffic flows is needed in order to assess the
whole impact of the new schemes.

3.3 Data
The individual labour market outcomes data comes from the Annual Survey of Hours and Earnings
(henceforth ASHE), which provides information about employees on an annual basis. The ASHE is
an annual survey of the earnings of employees in Great Britain, which from 2004 replaced the New
Earnings Survey (NES). Its primary purpose is to obtain information about the levels, distribution
and make-up of earnings, and for the collective agreements that cover them. It is designed to repres-
ent all categories of employees in businesses of all kinds and sizes. The questionnaire is directed to
the employer, who completes it on the basis of payroll records for the employee. The earnings, hours
of work and other information relate to a specified week in April of each year.
ASHE is based on a survey of a 1% sample of employees on the Inland Revenue PAYE register
(Pay As You Earn). The information is provided by the employer. The sample consists of employees
whose National Insurance numbers end with two specific digits. It covers approximately 160,000
individuals a year. The survey is designed as a panel of workers, in which the same workers are
observed for multiple years. The sample is replenished as workers leave the PAYE system (e.g. to
self employment, retirement, overseas or death) and new workers enter it (e.g. from school, self-
employment, immigration).
ASHE contains information on the make-up of weekly earnings and hours worked (basic, gross
and overtime), occupation (using Standard Occupational Classification – SOC), industrial sector (us-
ing Standard Industry Classification – SIC), collective agreement status, whether the job is private or
public sector, age, gender, postcode of workplace, and from 2002, postcode of residence. In order to
clean the data I drop the 0.5% top and bottom extreme values of the labour market outcome variables
(wages and hours) and of the commuting times. I define total pay consistently over the whole period
2002–200825 . I also removed observations with negative values of the variables and individuals which
show inconsistency in their age or gender over time. I only keep main jobs for those individuals that
have more than one job in the same year. Finally, I drop the individuals for which earnings were
affected by absence (loss of pay) and those paid at trainee/junior rates.
ASHE provides information on the occupation level of the individuals using SOC 2000 codes from
25 Different stratifications of ASHE define gross pay differently. I use the most recent definition: total (gross) pay=basic pay

+ incentive pay + shift and premium payments + overtime pay + other pay.

15
2002. I define broader occupation codes using the first digit of the code26 . I also define five broader
industrial categories based on 2-digit codes from the SIC 2003 classification27 .
The great advantage of this data is the good quality of the earnings and hours information and the
detailed information on the geographical location of both the workplace and the place of residence.
Furthermore, its panel structure allows us to control for unobservable time invariant characteristics
of the workers which might be correlated with the variable of interest. However, the survey contains
information only on workers who are employed, so I am unable to observe unemployment spells.
The responses to the survey are provided by employers, rather than the employees themselves, so
personal and household information about the employees is very limited (essentially gender and
age). However, most of the household characteristics change slowly over time so the use of individual
and individual-location fixed effect control for time-invariant heterogeneity captures most of these
characteristics.
Table A.2 provides summary statistics for the main variables of analysis for both panel defin-
itions. The table shows the mean, standard deviation, maximum value, minimum value for the
overall, between and within dimensions of the panels for basic weekly pay (wages), basic weekly
hours worked, accessibility from workplace, accessibility from home and predicted road travel time
between work and home, all in natural logarithms. The top panel displays statistics for the panel
defined using individual fixed-effects for the estimation sample (320,105 observations). We can see
there are 78,305 individuals which appear an average of 4.1 times. For all variables we can see that
most of the variance is between groups. The bottom panel displays the summary statistics for the
panel defined using individual-work-home fixed-effects. Even if the number of observations is smal-
ler28 , the number of groups is larger (79,868), and groups appear fewer times in the panel (3.35 times).
In this setup of the panel the concentration of the variance in the between dimension is even more
evident, especially for the accessibility and the travel time variables.

[INSERT TABLE A.2 HERE]

Apart from ASHE, other datasets are used to construct the accessibility measures (more inform-
ation is available in section A.1. in the online appendix). I use the National Statistics Postcode Dir-
ectory (NSPD), which provides a look-up between postcodes and higher UK geographies, to allocate
the individuals to work and home wards. This dataset also provides information about the number
of residential addresses in each postcode, which is used to calculate population-based effective dens-
ity measures. I also use data from the 2001 British Census to calculate socio-economic controls at the
ward level, which are included in the estimation of equation (5).

4 Results
4.1 Individual fixed effects
Table 3 presents the results for the estimates of equation (4) on log wages (log of basic weekly pay). In
this section I focus on the results on wages, while in sections 4.3.2 I explore the effects on other labour
market outcomes (mainly on labour supply). These results are obtained using the 30 kilometre band
both from work-ward and home-ward, and excluding the wards where the nearest scheme begins
26 They correspond to: 1 Managers and Senior Officials; 2 Professional Occupations; 3 Associate Professional and Tech-
nical Occupations; 4 Administrative and Secretarial Occupations; 5 Skilled Trades Occupations; 6 Personal Service Occu-
pations; 7 Sales and Customer Service Occupations; 8 Process, Plant and Machine Operatives; 9 Elementary Occupations.
27 I define 10 broad categories based on 2-digit SIC codes: 1 Primary Activities - agriculture, fishing, forestry, mining -

(SICs 1-14), 2 Manufacturing (SICs 15-37), 3 Energy & Construction (SICs 40-45), 4 Wholesale & Retail (SICs 50-52), 5 Hotels
& Restaurants (SIC 55), 6 Transport & Communications (SICs 60-64), 7 Financial Intermediation (SICs 65-67), 8 Real Estate
& Producer Services (SICs 70-74), 9 Public Administration, Education & Health (SICs 75-85), 10 Other Activities (SICs 90
and over).
28 If an individual now keeps a specific work-home combination only for one period, it drops from the panel estimation.

This is why the number of observations is smaller.

16
and ends. The table reports the estimated coefficients for log accessibility from both work ward, log
accessibility from home ward and log of predicted travel time between work and home wards. The
variables are constructed as defined above. The table reports the parameters and the standard errors
in square brackets. These are clustered at the individual level to allow for arbitrary correlation of
individual shocks over time29 . The different columns report results for different specifications where
different sets of controls are introduced. These are indicated in the last rows of the table. The sample
size is 320,105 observations, for 78,305 individuals observed an average of around 4 periods of time
(see table A.2). All observations include sic-by-year fixed-effect to control for nationwide industry-
specific shocks. The star symbols next to the estimated coefficient indicate the level at which the
coefficients are statistically different from zero.

[INSERT TABLE 3 HERE]

Column 1 reports the Ordinary Least Square results (OLS) where only sic-by-year dummies are
included. I use a standard definition of accessibility based on Art in columns (1) and (2). Given the
large number of observations, these partial correlations are precisely estimated. I find a positive and
significant relationship between accessibility from work on wages, a negative relationship between
accessibility from home and wages and a positive coefficient on the effect travel time on pay. The pos-
itive effect of accessibility from work on wages and hours could be due to agglomeration externalities
or to the fact that professionals, who earn more and work full time, are concentrated in work loca-
tions where accessibility is higher. At the same time, workers in low paid jobs such as basic services
could also be living in these locations and this could explain the negative coefficient of accessibility
from home on earnings and hours. Finally, these results suggest that longer commutes are capitalised
into higher wages as suggested for example by Manning (2003).
In column 2 I introduce individual fixed-effects to control for unobserved individual heterogen-
eity that might be correlated with accessibility levels and wage levels. I now estimate the effects us-
ing the within-individual variation and exploiting the changes in accessibility over individual means
over time. The coefficient for accessibility from workplace is reduced substantially, and that of access-
ibility from home becomes positive. This suggests that individuals which have unobservable char-
acteristics correlated negatively with wages are located in wards in which accessibility from home
is higher. This is in line with the previous argument that less skilled workers live in denser areas in
which they can access more jobs easily, as predicted for example by the spatial mismatch hypothesis.
In column 3 I use Âiht , the measure of accessibility that keeps ward employment fixed to 2001
levels. I do this in order to avoid using the variation which comes from spatial relocation of em-
ployment, as these changes and individual wages are probably influenced by the same unobservable
factors. The coefficient barely moves, which suggests that both variables are highly correlated and
that the instrument is very strong. In column 4 I control for initial accessibility (both for work and
home) and initial job and personal characteristics, and the coefficients remain unchanged. In column
5 I add (closest) scheme dummies, distance to the scheme and opened scheme dummies to control
for specific unobservables around the scheme. In column 6 I add 2001 census controls and commute
band dummies. This is my preferred specification as it includes all the individual, job and area level
controls. The coefficient of accessibility from workplace is 0.038, that of accessibility from home is
0.021 and the effect of predicted road commute time is 0.014. The are all significantly different from
zero at 1% level. These elasticities are fairly small, suggesting that doubling accessibility at workplace
(what, given the summary statistics in 1 would require large amounts of new links) would result in
an increase in wages of around 4%, and increasing accessibility from workplace of around 2%. Even
if small, the estimates of the effect of accessibility from work are similar in magnitude to the estimates
29 Alternatively I could clustered the s.e. at ward levels, but as individuals move over space, these levels are not nested.

However, when we allow for spatial mobility the relevant clustering level might be the individual, not the area, as he is
choosing the “level of treatment” by his moving decisions. In practice, the main results rely on fixed-location estimates
from equation (5), where I am able to cluster at the area treatment level (the pair work ward-home ward location).

17
of the effect of market potential on wages provided for example by Combes et al. (2008) or Mion &
Naticchioni (2009).
In columns 7 and 8 I provide additional estimates for robustness. In column 7 I estimate the same
specification as in 6 but using instrumental variables, and the coefficients change little. The F-statistic
of the first-stage (Kleibergen-Paap rk Wald F statistic) indicates that the instrument is very strong.
This is not surprising for two reasons; the first one is that Âiht and Aiht share one source of variation,
changes in predicted travel times, so there is a mechanical relationship between the two. The second
one is that, as shown in Gibbons et al. (2012), new roads attract employment to areas around where
they are placed, so the changes in travel times are good predictors of changes in local employment.
The second additional result, presented in column 8, uses the controls no only in levels (initial values)
but also interacted with a linear trend (when possible). The results barely change with respect to the
main results in column 6.
As discussed above, the identification of the three coefficient relies on the individual variation
stemming from spatial relocations. This implies that individuals might be changing jobs or home
as a result of changes in accessibility or due to unobservables correlated with these changes. In the
next section I investigate if using the variation of accessibility for individuals while they keep their
location fixed has any impact on the estimates of the effects.

4.2 Exploring the role of spatial mobility

This section explores the impact that restricting spatial mobility has on the estimates of the effect of
accessibility changes on wages. I do this using two methods. The first one is restricting the sample
used in the estimates of table 3 to individuals that do not move home ward, work ward or both during
the period of observation. This implies dropping observations from the sample, and running the
results on individuals that meet the stable location conditions. The second approach uses individual-
location fixed effects. As explained above, this has the advantage of not “dropping” individuals but
it allocates different individual fixed-effects depending on their work-home location30 . Individuals
might change locations over time but, conditional on having at least two observations for a given
location choice, we use the information for the whole spell in which in which the worker is observed.
Table 4 presents these results.

[INSERT TABLE 4 HERE]

The dependent variable in this table is log of basic weekly pay. All the columns include all controls
either in dummies or in trends, depending on the specification. Column 1 replicates the main result of
table 3 for comparison. In this specification individuals are allowed to change work and home wards
at any point in time. In columns 2-4 I restrict the sample to workers which keep stable locations
(either work or home, or both) over the period of study. Column 2 drops the individuals that change
homes at least once during the period of analysis. Almost 90,000 observations drop. The coefficient
of the effect of accessibility from work changes a little with respect to column 1, and the same for
the effect of travel time. However, the coefficient of the effect of accessibility from home become
insignificant. This is not due to changes in the size of the coefficient (in fact, this increases), but to the
fact that standard errors increase dramatically when we restrict the sample to non-home-movers. This
suggests that most of the identification of the effect of accessibility from home on wages in column
1 relies on variation stemming from spatial home relocation. Even if this strategy does not directly
test if workers move as a consequence of road construction, it suggests that they move to areas in
which accessibility is increasing due to road construction31 . This means that residential sorting is a
30 Of course, the number of observations in both panels is different because we need individuals to keep a stable location

for at least two periods in order to be able exploit the within individual-location variation. When we use individual fixed-
effects individuals need to appear more than one but they can be located in different wards.
31 Recall that accessibility changes are non-negative as in the calculation of Â
iht ward employment is fixed to 2001 levels

18
consequence of new links.
In column 3 I restrict the sample to workers that do not change work ward during the period of
study, but they can change home location. The sample is even smaller than before, around 190,000
observations as compared to 320,000. The coefficient of accessibility for workplace remains significant
at 1%, but both the size of the parameter and the standard errors increase. The estimate become less
precise because there is less variation from which to estimate the effect, given that workers keep
stable work wards in this sample. The coefficient of predicted travel time on wages is also significant
and positive. In column 4, I restrict the sample to individuals that do not move either work or home
ward during the period of analysis. The sample in which I estimate this results is substantially smaller
than the original one, around 150,000 individuals. The coefficients are very similar to those of column
3. However, the coefficient of the effect of travel time collapse to zero which suggests that workers get
compensated for their commute when they change jobs, and for a given job, changes in travel time
due to road construction do not have a direct effect on pay. The standard errors for all 3 variables
are larger given that I have less variation in both accessibility and travel times when work and home
locations are stable over time. In both columns 3 and 4 only accessibility from workplace is significant,
an the elasticity is around 2.5%.
The difference in size of the coefficient of the effect of accessibility from workplace obtained in
column 1 and those obtained in column 3 or 4 is quite substantial. Specifically, the reduced-form
estimates of column 4 are almost one order of magnitude larger than the ones using individual fixed-
effects. The difference in the size of the coefficient originates from keeping work ward locations for
individuals fixed over time, regardless of home mobility. This would suggest that individuals capit-
alise any benefits that firms draw from increases in effective density to a larger extent when they stay
in the same job-area for a period of time.
For simplicity, we can separate the expected effect of accessibility from workplace on wages in the
effect due to job mobility (better job matches) and that due to changes in density at current workplace
(related to agglomeration economies in-situ). If the channels through which increases in effective
density are positively affecting wages are linked to area specific agglomeration economies (for ex-
ample knowledge spillovers or labour market pooling), it is sensible to assume that workers would
increasingly capitalise these benefits over time while staying at the same work location. In these spe-
cification we are keeping work ward fixed, so workers could still change jobs within their ward (this
is explored in section 4.3.1). Recent evidence suggests that workers benefit from better job matches in
larger labour markets (Harmon, 2013), but my finding suggests that the effect of changes in the size
of labour markets on wages works mostly via location-specific channels, potentially agglomeration
economies. Moreover, recent evidence from De la Roca & Puga (2012) suggests that workers in more
agglomerated areas do not necessarily have higher initial ability, but the effect of density in wages
increases with workers’ experience in agglomerated areas.
The difference in the coefficients between column 1 and columns 4 could also be due to sample
selection, as I drop the observations which do not meet the stable location criteria in the different
columns. To check the extent to which this is an issue, in table A.3 I provide some summary statistics
on the initial values for some variables for the whole sample and for individuals that move any of the
two dimensions (movers) or those that keep stable work-home locations over time (stayers). We can
observe that these differences are minor and not statistically significant, especially if we compare the
stayers and the whole sample. Therefore, the difference in the size of the coefficients is likely due to
the effect that restricting mobility has on the source of variation for the accessibility and travel time
variables.
In columns 5 to 7 of Table 4 I use individual-location specific fixed-effects. By including worker-
ward specific fixed-effects I use a larger number of observations and exploit the variation for the same
individual for different locations. The coefficients are very similar to those of columns 2 to 4 due to
and changes in optimal travel times between locations are non-negative. See the online appendix for more detail.

19
the absence of sample selection bias in the estimates as discussed above. In my preferred specifica-
tion, that of column 7, the elasticity of accessibility from workplace on wages is around 3.1%, and
there is no effect of accessibility from home nor predicted commute travel time. This coefficient im-
plies that doubling accessibility from work, for a given work-home location, would increase weekly
wages around 3%. This effect is not negligible but it is fairly small. Even if this coefficient is very
similar to that of column 4, I prefer this specification because it uses a larger range of variation and a
larger number of observations. Therefore, the robustness checks of next section are performed with
reference to these coefficients.

4.3 Additional results and robustness

4.3.1 Robustness checks

The main result from the previous section is that, once we control for spatial mobility, only access-
ibility from workplace has a positive significant effect on workers’ wages. In this section I test the
robustness of these findings. Only the coefficients for work accessibility, home accessibility and pre-
dicted travel time are reported in the tables. Unless specified, all the results are obtained using the
30-kilometre band (excluding the extremes) and include all the controls and trends. The main results
are reproduced in the tables to ease comparison.
In order to link the empirical results to the theoretical predictions, in equation (5) I include access-
ibility from both work and home locations and predicted commuting time. But these three variables
might be very correlated, threatening the correct identification of the parameters. For this reason,
in columns 2 to 5 in table 5 I test the robustness of the findings to changes in the specification. We
can see that the effect of work accessibility remains significant even if when I exclude travel time or
home accessibility for the specification, and that by itself, home accessibility remains insignificant.
This rules out the possibility that, due to collinearity, work accessibility would be capturing all the
effect of effective density for workers which have work and home locations close to each other. In
column 6 I exclude observations for which work and home wards is the same (when both accessibil-
ity measures would be overlapping). The main result remains mostly unchanged. In column 7 I use
one lag of the accessibility measures and still find a positive significant impact of accessibility from
work on wages. Finally, in column 8 I use the accessibility indices as defined in Art and instrument
them with Âiht . The coefficient is very similar to that of column 1.

[INSERT TABLE 5 HERE]

In the main results, the individual-location fixed effects are defined using work and home ward
locations. Wards are very small units and the incidence of house and job changes within these small
units is very small and unlikely to affect the main results. However, changes in the size of the coef-
ficient when we add additional mobility restrictions points towards the channels through which
accessibility is affecting wages. Table 6 explores this. Workers’ home postcode are available in ASHE
and, given that in Great Britain postcodes refer roughly to house blocks, we can track individuals
almost at their address level. For job locations, ASHE does not provide information on plant iden-
tifiers, but it provides information on work location postcodes, firm sector and firm identifier. I use
a combination of these three variables to define specific plant locations. In column 2 of table 6 I use
individual-home postcode-plant location effects, which adds additional mobility restrictions with
respect to the main specification of column 1. The qualitative results remain unchanged: only access-
ibility from workplace has an effect on wages and the elasticity is now equal to 2.5%. Columns 3
and 4 investigate if this change in the coefficient is due to house or plant moves within the wards. In
column 3 I use individual-home postcode-work ward fixed-effects: this implies that individuals keep
the same house and work wards but are allowed to change plants within the ward. The coefficient
is larger than in column 2 and more significant, but smaller than that of column 1. In column 4 I

20
use individual-home ward-plant specific fixed-effects, so I exploit variation in accessibility can still
stem from changes in home postcodes within the wards. The result is very similar to that of column
3. The fact that the coefficient is larger when we allow for in-ward job movements suggests that
one of the channels through which effective density is being capitalised into wages is by very local
plant moves. There are two plausible explanations for this: firms located in wards which experience
changes in accessibility are experiencing productivity shocks and are able to pay more, and work-
ers could take advantage of this by moving to other plants close-by and capitalising these gains via
higher wages. Another explanation would be that labour markets are effectively larger due to faster
road connections now and firms located close to the schemes start competing for specialised labour
with a larger pool of firms. In order to keep their workers, they bid their wages up. This would imply
that employed workers market-power increases due to the road construction. These explanations are
consistent with the findings in Gibbons et al. (2012).

[INSERT TABLE 6 HERE]

A potential concern for the validity of the results would be that the is a lot of overlap between
the work 30-kilometre band and the home 30-kilometre band, so we are unable to separate the ef-
fects of work and home accessibility and that is why I only find significant impacts from one of
the dimensions (work). A first attempt to control for this is to include commute distance trends in
the estimations, as I do in the main results. I explore this further in table 7. I restrict the sample to
individuals for whom work and home locations are located at different distances.
Given that the bands around the schemes are defined using crow-fly distance, I do the same
here to restrict for commuting distances so it is a better control for the overlap of the work and
home distance bands. I use restrictions for 30, 20, 15 and 10 kilometres32 . When these distances are
small the overlap between the work and home distance bands would be large, while when they are
larger, there would be less overlap between the work and home areas. Column 1 reproduces the main
results. In columns 2 to 5 I use different work-home distance restrictions and the main conclusions
remain unchanged. The coefficients change a bit with respect to column 1, but they are very similar
and still significantly positive. In columns 6 to 10 I restrict the observations to workers for which
the closest scheme is the same for work and home locations. This is not necessarily the case in the
previous results, as depending on the location of the schemes and the work and home wards workers
might be closer to different schemes in the two dimensions. The results in columns 6 to 10 are very
similar to before.

[INSERT TABLE 7 HERE]

Tables A.4 and A.5 check the robustness of the findings to excluding different wards and using
different size of the distance bands around work and home locations. Column 2 replicates the main
results. The coefficient on accessibility from workplace changes very little when I do not exclude any
or when I exclude more wards at the beginning and end of the schemes. Column 6 is a bit different. I
exclude not only the wards at the extreme of the schemes but also the wards along which the scheme
goes through. This strategy is similar to that used by Faber (2012). The coefficient is smaller but less
significant (only at 10%). This suggests that most of the effect is drawn from locations very close to the
schemes but not necessarily aimed at them. This indicates that the effects are very localised. In table
A.5 I expand (columns 1 to 4) and reduce (columns 6 to 10) the size of the work and home distance
bands. The coefficients are very similar across the columns only becoming insignificant when the size
of the bands is very small and the number of observations very reduced.
In table A.6 I use alternative definitions of the accessibility index Âiht . Column 1 reproduces the
main results where I use ward employment in 2001 as the measure of economic size of the wards.
32 The average distance between work and home locations in the individual-location panel is 9 kilometres, the 25th per-

centile is 2.3 kilometres, the median is 5.3 kilometres and the 75th percentile is 11.7 kilometres.

21
Columns 2 and 3 use different measures: 2001 number of plants and 2001 residential address counts
(a proxy for population). The coefficients are very similar to the main results. Columns 4 to 6 use
ward 2001 employment but change the form of the cost function and/or the degree of the distance
decay. As expected the size of coefficients is different, as the weights given to the different wards in
the computation of the accessibility indices change. But the qualitative results remain unchanged.
In table A.7 I exclude observations of workers located in London33 . In column 1 I replicate the
main results. In column 2 I exclude individuals working in London at least once during the panel, in
column 3 individuals living in London and in column 4 individuals living or working in London at
least once while observed. The results remain very similar to the main findings.
Finally, In table A.8 I explore the effect of using individual-location fixed-effects further. The main
results the observations for workers irrespectively of their work-home location, I just allocate a new
fixed-effect when a new location-pair is chosen. Of this sample, there will be some individual that
always keep the same location (in both or one of the dimensions) so the individual-location fixed-
effects would be equivalent to individual fixed-effects. Other individuals would have more than one
spell, meaning the get different sets of individual-location fixed-effects over time. In columns 2 to
3 I restrict the sample to those who keep the same work-home location over time (column 2), keep
the same work location (column 3) or keep the same home location (column 4). The results remain
very similar to the main findings. In columns 5 to 7 I focus on individuals that get more than one
individual-location fixed effect. The samples are quite small, and yet I still get very similar results,
especially for column 5.

4.3.2 Other labour market outcomes

In this section I explore the impact of changes in accessibility in other labour market outcomes. In
particular, I study the effect of changes in effective density due to road construction on labour supply
(hours and full-time status) and hourly earnings. I use the same identification strategy than in the
previous sections.
Changes in effective density might affect hours worked via several channels. If wages are react-
ing as a consequence of road construction, workers’ might have an incentive to work longer hours
in order to gain more. Alternatively, if labour markets become more competitive because better con-
nections are making then tighter (Détang-Dessendre & Gaigné, 2009), for a given commuting cost
workers might work more in order to keep their jobs or to establish themselves in the labour markets
(Rosenthal & Strange, 2008). However, number of hours worked per week might not be a flexible
outcome to adjust given a contractual setup. For this reason, I explore if workers are more likely to
work part-time or full-time as a consequence of the road investments. As male and female labour
supply might react differently, I also study the differences by gender for this variables.
Table 8 presents the results for specification (5) using different labour market outcomes as de-
pendent variables. The outcomes can be divided into basic and gross/total (including overpay and
overtime). Most of the outcomes are weekly, but I also test the effect for total annual pay which
might include additional extras from for example bonuses. Columns 1 to 3 display the results for the
wage variables; using gross wages or annual wages does not cause large changes in the coefficients.
Columns 4 and 5 test the effect of changes in accessibility on weekly hours worked, both basic and
total. As for pay, only accessibility from work has an effect on labour supply, and the coefficient is
slightly larger for total hours (even thought not statistically significant from the one on basic hours).
Columns 6 and 7 study the impact on hourly earnings, which are defined as the weekly pay over
weekly hours. Focusing on the effects on basic outcomes, the elasticity of wages with respect to work
accessibility is larger than the one of hours. However, when I test the effect on the ratio of both vari-
ables, basic hourly earnings, the effect is weaker but still significant and positive. The elasticity is
33 Defined using the local labour market definition of London as in Gibbons et al. (2010)

22
lower than in the case of weekly wages, around 1.1%, which is consistent with the elasticity of pay
being larger than that of hours. For total hourly earnings, the coefficient is non-significant.

[INSERT TABLE 8 HERE]

We would expect that increasing access to jobs would lower barriers to participation and have an
effect especially on women, for example if they commute shorter distances than men (as documented
for example by Rosenthal & Strange, 2012). I cannot directly test this in my data, but I can explore if
there are gender differences on labour supply for employed women. I do this in table 9. In column
1, I test if accessibility has an effect on working full-time (versus working-part time). This result is
reported in column 1; I find a positive effect of accessibility from work on working-full time, but
negative significant effects of accessibility from home and commute travel time on this outcome. As
discussed above, as it is mainly full-time workers who benefit from the accessibility-induce wage in-
creases, this could results in workers moving from part time to full time status while remaining in the
same work-home locations. Commute distance and labour supply can be positively related because,
the longer the commute is, the more worthy is to work longer hours in order to compensate for the
time spent travelling to work. There is no straightforward explanation why better home accessibility
could increase the probability of working full time, given that this estimates are obtained for fixed
work-home locations.
It could be that there are substantial gender differences with respect to this margin of adjustment
when road construction takes place. I explore this in the remaining columns of table 9. Columns 2
and 3 test the effect of accessibility and travel time on full time status separately for male and female
workers. I only find effects for female workers, and only for accessibility from workplace (positive)
and commute time (negative). The impact of work accessibility is also only significant for women
when I test the impact on basic hours worked (columns 4 and 5). Only female workers seem to be
working longer as a result in increases in work accessibility and reductions in commute travel time
(as suggested by Black et al., 2013). This suggests that women are more likely to adjust their labour
supply when wages increase as a result of changes in work accessibility. As I report in the previous
section, the accessibility wage premium is larger for women, which could explain the result.

[INSERT TABLE 9 HERE]

Section A.2.1. in the online appendix explores the heterogeneity of the effects.

4.4 Discussion
Section 2 discusses the multiple channels through which changes in effective density via the con-
struction of new road links can affect individual labour market outcomes, especially wages.
For individuals working and living in different locations, I test the impact of accessibility from
workplace and accessibility from home, and also the impact of changes in predicted travel time
between work and home on wages. We could explain the impact from workplace as a result of ag-
glomeration economies. Firms might become more productive due to the transport investments (as
suggested by Gibbons et al., 2012) and workers might be able to capitalise these gains, especially if
they have market power over firms when specialised labour is scarce. Changes in accessibility from
home could have an impact on wages if workers get better labour market matches as a result of in-
creased density. Finally, lower commute times after road construction could be also capitalised into
higher wages if workers become more productive when they have to spend less time traveling to
work (van Ommeren & Gutiérrez-i-Puigarnau, 2011).
Results in table 3 provide the estimates when I use individual fixed-effects and I use the variation
in the three variables for an individual irrespective of his location. All three coefficients are significant
and positive. Given the summary statistics reported in table 1, between 2002 and 2008 the average
increase in wages due to increases in accessibility from workplace would be 0.012%, and that from

23
home would be 0.0064%. When we get close to the site of the new links, these effects increase (up to
0.077% and 0.043% within 10 kilometres), but they are still small. The size of the coefficient of work
accessibility is similar in nature to the estimates of the impact of “market-access” (or market potential)
on wages (some examples are Mion & Naticchioni, 2005; Fingleton, 2006; Combes et al., 2008; Amiti
& Cameron, 2007). In their review of the literature, Graham & Melo (2009) find that these estimates
are normally between 0.02 and 0.2. Even if my estimate are not directly comparable as my definition
of market-access only relies on variation due changes in optimal travel times, the coefficient for work
accessibility falls within that range. In fact, in column 7 of table 3, when I also use the variation in
access stemming from changes in local employment, the coefficients are only marginally larger.
The estimates using individual fixed-effects rely on variation stemming from spatial relocation
of workers across work and home wards. As these moves could be due to road construction, or to
unobservables related to them, in table 4 I use two methods to explore the impact of mobility on the
estimates. I either restrict the samples using mobility restrictions or I use individual-location specific
fixed-effects. The coefficient using both methods are very similar. The results change substantially
with respect to table 3; the coefficient of accessibility from home becomes insignificant while the
coefficient of accessibility from workplace remains significant and increases significantly. At the same
time, the coefficient of home accessibility becomes insignificant when using individual variation for
a given location choice. This indicates that the variation in home accessibility stemming from home
mobility is essential to be able to identify the effects of accessibility from home on wages. This sug-
gests that home sorting is an outcome of the reduction of travel times. Given the construction of the
travel times, the changes I exploit are always non-negative, which implies that workers are moving
to locations where accessibility is increasing.
Once individuals have chosen their residential location, access to more jobs from home does not
seem to have an effect on their wages neither on hours, conditional on predicted commuting time.
The theoretical channel through which better accessibility from home would have an effect on wages
once controlling for sorting is unclear. The spatial mismatch hypothesis predicts positive effects of
better access to jobs from home on labour market outcomes, especially on the probability of becom-
ing employed or on the length of the unemployment spells. Better home access can boost wages if
workers can find better job-matches when changing jobs. However, from the results obtained using
the sample, in which all the individuals are already working and keep their work-home location
fixed, I do not find any evidence in favour of the spatial mismatch hypothesis. This could be due to
households having already optimised their residential location in order to have access to better jobs
and shorter commutes so further increases in accessibility from home do not have any impact on
their wages or hours worked.
When I use individual-work-home fixed-effects the coefficient of accessibility from work on wages
increases largely, almost tenfold. Spatial sorting could be explaining part of the difference in size
between my estimates and the coefficients found in previous evidence. The identification strategy I
follow substantially helps to reduce the bias caused by the spatial sorting of workers. As explained
above, I control for sorting at the individual level (both work and residential sorting) by using the
individual-work-home fixed effects. This way, for the identification of the effects I exploit the changes
in accessibility over time for an individual while staying a given location combination. Following the
discussion above, selection of workers does not seem a major issue, so the difference in the estimates
when I restrict spatial mobility are likely due to the difference source of variation used when I exploit
the variations in accessibility in-situ. This way, the effects of agglomeration on wages would work
“while workers stay in the same location”. Moreover, as outlined above, the instrumental variables
strategy (using Ârt ) helps to eliminate the bias induced by spatial relocation of workers across space
which might be driven by the changes in accessibility. In other words, I tackle the sorting of workers
at the ward level and avoid using these endogenous variation in the estimation of the effects34 .
34 In their investigation of the determinants of individual wages using a large sample of French workers, Combes et al.

24
 The positive effect of accessibility from work on wages and hours, given that in tables 3 and 4 I
control for commuting travel time, could be working through spatial competition or agglomeration
externalities. If firms in which the workers are employed can access a larger pool of workers, then
employees might behave more competitively and work harder (especially high skilled workers, as
suggested by Rosenthal & Strange, 2008). Workers could become more productive in these denser
and better connected areas, as has been suggested and empirically verified in the agglomeration
economies literature (see Duranton & Puga, 2004, for a review) or in the new economic geography
literature (Krugman & Venables, 1995; Redding & Venables, 2004, for example). Given that the effects
are obtained for a given work-home location, the channels is work-area or firm-area specific. Faster
travel times increase the effective size of the labour markets (Harmon, 2013). If firms compete for spe-
cialised workers in the local area and competition for these workers increases with effective density
changes, firms might have to pay workers higher wages in order to keep them in the firm.
In my preferred specification, when I use individual-work-home specific fixed-effects, the coeffi-
cient of work accessibility is equal to 0.309. Given the average growth in accessibility between 2002
and 2008, almost 0.1% of the growth in wages could be attributed to road construction. This is a small
effect but significantly different from zero. As we get closer to the new links (10 kms) it amounts to al-
most 0.63%. This effect is larger around schemes which had larger impacts on predicted travel times.
Within 10 kilometres of the A5 Nesscliffe Bypass, opened in 2003 and which was almost 21.5 kms
long (see table A.13), the average change in accessibility between 2002 and 2008 was 3.62%, which
implies a wage growth of around 1.12%. Around the A1(M) Ferrybridge to Hook Moor (19.2 kms and
opened in 2006), the effect on wages is almost 1%. Given that the average growth in basic weekly pay
for the whole Great Britain between 2002 and 2008 was around 25%, the magnitude of the impact of
roads on wages is rather limited. For the whole Great Britain the size of the coefficients implies that
road construction explains around 0.4% of total wage growth during this period. As we get close to
the schemes, the effects become more important.
One of the main contributions of the paper is to provide robust estimates on the impact market
potential on wages using a different strategy from most of the existing evidence and focusing on a dif-
ferent channel. The papers mentioned above use a general market potential definition (Harris, 1954),
which is similar in structure to (2) but uses geodesic time invariant distance as the measure of prox-
imity between locations. As the distance between locations (weights) is fixed over time, the effects
of market potential on wages is estimated using changes in employment in the different locations.
The endogeneity of these changes is dealt in different ways, for example constructing instruments
with historical or geological data (see for example Ciccone & Hall, 1996; Mion & Naticchioni, 2009;
Combes et al., 2011). In the current paper, the main results are obtained using a variation of (2) which
keeps employment constant and changes over time only due to reductions on optimal travel times
between locations. This strategy focuses on changes induced by reduction on travel times after new
links are added to the road network. As discussed above, it also avoids using endogenous variation
in Ar t from spatial changes in ward employment.

5 Conclusion
In this paper I investigate the effect of changes in accessibility, induced by road construction, on
individual labour market outcomes. I make use of rich individual datasets and small geographical
(2008) find that differences in the skill composition of the labour force account for 40 to 50% of aggregate spatial wage
disparities. They conclude that workers with better labour market characteristics tend to agglomerate in the larger, denser
and more skilled local labour market. The intuition is that sorting of workers across space has an effect on individual wages
because more skilled workers sort in specific areas. My approach is different as I explicitly control for the work and home
sorting of the individuals. Furthermore, I focus on accessibility changes stemming from road construction and do not use
the variation coming from spatial changes in employment. This should help to reduce sorting issues in the same line as
these authors.

25
scale to be able to infer causality on the estimates.
I provide evidence of the effect of accessibility both from work and home wards on individual
wages, and also on labour supply. Controlling for commuting time allows us to learn about the poten-
tial theoretical channels through which accessibility might be impacting on labour market outcomes.
The methodology used overcomes several endogeneity problems common in the identification of
effects of access to jobs: unobserved individual heterogeneity, reverse causation between accessibil-
ity and spatial distribution of local employment and endogenous placement of the new road links.
Moreover, I explore the impact that spatial mobility (sorting) has on the significance and size of the
estimates. When I exploit the changes in accessibility for a given work-home location, I find positive
effects of accessibility from work on earnings and total hours. These results could be driven by ag-
glomeration externalities which are capitalised into higher nominal wages and by increased spatial
competition which might make employees work longer hours. Workers seem also to be responding
to higher nominal wages by switching from part time to full time jobs. These results are very robust
to different specifications and across different groups of workers.
Two main conclusion can be drawn from these results. Firstly, my identification strategy separ-
ates the effect of spatial sorting from other channels through which accessibility might be impacting
labour market outcomes. As discussed in the text, once I focus on the variation of accessibility for
a given work-home location combination, there is no effect of accessibility from home on wages
or hours. This suggest that workers sort residential location in response to changes in accessibility.
On the other hand, the effect of accessibility from workplace is significant even once controlling for
sorting, and the size of the coefficient increases substantially. This result stresses the importance of ac-
counting for sorting to be able to identify the channels through which accessibility is affecting wages.
My findings add to the evidence of the importance of learning on the determination density wage
premium (see Baum-Snow & Pavan, 2012; De la Roca & Puga, 2012, for recent evidence). Given that
I exploit the variation on effective density only via changes in travel times (keeping initial ward em-
ployment fixed) and for a given work-home location choice, my estimates do not capture the effect
that spatial sorting of workers (both at the aggregated and the individual level) has on the effect of
density on wages (Combes et al., 2008). Moreover, the results suggest that learning effect might be
not only location-specific but even job-specific.
The main results estimate the effect of accessibility changes on wages, but I also explore the im-
pact on other labour market outcomes, namely hourly earnings and hours worked. I find effect of
accessibility from workplace on all three variables, suggesting that workers wage gains are a com-
bination of longer working weeks and increased hourly pay.
The findings in this paper provide new evidence on the effect of agglomeration externalities and
proximity to markets on individual wages and hours worked using a novel strategy that carefully
tackles multiple endogeneity issues and a detailed dataset on road construction. I use changes in
optimal travel times between location stemming from road construction as the source of changes in
accessibility. Road construction can be directly used by policy makers to influence travel times and
therefore affect effective density and proximity between locations. Transport policy is a substantial
part of economic policy and the estimates of this paper help to shed light on the economic impacts
that transport infrastructure investments can have on individual labour market outcomes.

26
Tables and figures

C HANGE LOG ACCESSIBILITY BETWEEN 2002 AND 2008

Standard Minimum 90th Maximum Proportion
Wards Mean Median
deviation value percentile value of zeroes
All 10540 0.305% 1.634% 0.000% 0.004% 0.572% 52.368% 42.884%
30 kms 4130 0.749% 2.546% 0.000% 0.122% 1.570% 52.368% 10.823%
20 kms 2751 1.042% 3.069% 0.000% 0.223% 2.209% 52.368% 10.978%
10 kms 1119 2.024% 4.592% 0.000% 0.556% 4.708% 52.368% 9.294%
5 kms 539 3.445% 6.208% 0.000% 1.260% 8.504% 52.368% 5.937%
A NNUAL LOG ACCESSIBILITY 2002–2008
Overall Between Within Minimum 90th Maximum
Wards Mean Median
std. dev. std. dev. std. dev. value percentile value
All 73780 14.705 1.172 1.172 0.007 5.538 14.875 16.010 19.316
30 kms 28910 15.234 0.807 0.807 0.011 10.312 15.324 16.164 18.193
20 kms 19257 15.166 0.739 0.739 0.013 11.505 15.290 15.952 18.166
10 kms 7833 15.030 0.750 0.750 0.020 11.505 15.195 15.778 17.621
5kms 3773 14.970 0.755 0.755 0.028 12.075 15.156 15.705 17.082
Source: Department of Transport, BSD and own author’s own calculations. Index calculated using 2001 ward employment
and 2002-2008 predicted travel times. Top panel in percentage values.

Table 1: Summary statistics ward log of employment accessibility. Based on Gibbons et al. (2012)

Balancing tests: 2001 work ward values 30 kms band

Log of Log of Log of Log of Log of Log of
D EPENDENT VARIABLE : employment weekly hourly total commute commute
accessibility wages rate hours distance travel time
Log distance to closest 0.005 0.014 0.005 0.002 -0.01 -0.031*
scheme [0.010] [0.009] [0.004] [0.005] [0.020] [0.017]
Observations 4,037 3,627 3,624 3,624 3,761 3,749
Scheme dummies Y Y Y Y Y Y
Census controls Y Y Y Y Y Y
Notes: S.e. clustered at ward level. Wards up to 30 kms to the new schemes, exclude those located in the extremes
of the schemes. The regressions include 2001 census ward controls (mean age, unemployment rate, proportion of
workforce with higher education, proportion of population living in social housing, ward address density, proportion
of commutes by motor vehicles, average commuting distance) and closest scheme dummies. Commute distance and
travel time values correspond to 2002, the rest to 2001 values. Results for home ward very similar. Sources: ONS, DfT,
CASWEB and author’s own calculations. * p<0.1, ** p<0.05, *** p<0.01.

Table 2: Balancing tests: work ward values for 2001 (30 kms band). Based on Gibbons et al. (2012)

27
 ! !

Aberdeen Aberdeen
! !

!
Glasgow !
!
Glasgow !

Newcastle
!
Newcastle
!

A6
6

Leeds Bradford Leeds

A1(

! ! ! !
M

Liverpool Bradford
)

!
!
!
Liverpool
!

Manchester

!
Notingha m !
Notingha m
M
A

6( T Leicester
5

) A47
!
Birmingham !
A6

! !

Birmingham Cambridge Cambridge

A6

! !
43
A

Oxford ! !
Oxford
Cardiff Bristol !
London Cardiff Bristol !
London
! !
! !

2008 Major Road Network, type and location of new projects Change in log accessibility between 2002 and 2008
New routes Faster routes 0.00% - 0.05% 0.0501% - 0.50% 0.501% - 5.00% 5.01% - 52.37%

(a) Major road network 2008 and new roads projects (b) Change in log accessibility (2001 employment)
A1 (M )
A6
50

Leeds
Bradford ! A6 3
!
A1 (M )

Manchester
!

Change in log accessibility between 2002 and 2008 0 5 10 20 30 40

Kil om et ers
0.00% - 0.05% 0.051% - 0.50% 0.501% - 5.00% 5.001% - 52.37%

(c) Change in accessibility in Manchester-Leeds area (2001 employment)

Figure 1: Location of projects and changes in log accessibility 2002–08. Based on Gibbons et al. (2012)

28
D EPVAR : Log of basic weekly pay
1 2 3 4 5 6 7 8
Log of accessibility 0.206*** 0.071*** 0.073*** 0.073*** 0.065*** 0.038*** 0.044*** 0.034***
from work ward [0.005] [0.007] [0.007] [0.007] [0.007] [0.008] [0.008] [0.008]
Log of accessibility -0.065*** 0.039*** 0.041*** 0.041*** 0.031*** 0.021*** 0.024*** 0.016**
from home ward [0.006] [0.007] [0.007] [0.007] [0.007] [0.008] [0.008] [0.008]
Log of travel time 0.130*** 0.029*** 0.029*** 0.029*** 0.029*** 0.014*** 0.028*** 0.012***
between work & home [0.002] [0.002] [0.002] [0.002] [0.002] [0.002] [0.002] [0.002]
Observations 320105 320105 320105 320105 320105 320105 320105 320105
Individual FX N Y Y Y Y Y Y Y
Reduced form access N N Y Y Y Y Y Y
Initial access N N N Y Y Y Y Y
Job-perso characteristics N N N Y Y Y Y Y
Scheme dummies N N N N Y Y Y Y
Distance/open scheme N N N N Y Y Y Y
2001 Census dummies N N N N N Y Y Y
Commute bands dummies N N N N N Y Y Y
Instrumental variables N N N N N N Y N
Add controls in trends N N N N N N N Y
Notes: DEPVAR stands for dependent variable. Clustered standard errors shown in brackets. Distance band 30 kms (excluding extreme
wards). Personal and job characteristics include log of firm size (employment) and female, initial 1-dig occupation, initial age group, and
full-time dummies interacted with a linear trend. Ward 2001 attributes include mean age of the population, proportion of household
living in social housing, proportion of household with higher qualifications, unemployment rate, address density, average distance
traveled to place of work and proportion of employees going to work by motor vehicles. All specification include sector-year dummies
(10 categories). F-stat first stage in column 7 is equal to 297,528. “Ind” stands for individual. S.e. clustering (individual). * p<0.1, **
p<0.05, *** p<0.01. Source: ONS, DfT, CASWEB and author’s own calculations.

Table 3: Individual fixed effects results, log of basic weekly pay

D EPVAR : Log of basic weekly pay

Indiv FX Indiv FX + restrictions Indiv-ward FX
1 2 3 4 5 6 7
Log of accessibility 0.038*** 0.045*** 0.244*** 0.261*** 0.039*** 0.246*** 0.309***
from work ward [0.008] [0.010] [0.087] [0.080] [0.008] [0.090] [0.096]
Log of accessibility 0.021*** 0.048 -0.008 -0.002 0.098 -0.004 -0.058
from home ward [0.008] [0.111] [0.009] [0.086] [0.101] [0.007] [0.080]
Log of travel time 0.014*** 0.017*** 0.000 -0.006 0.017*** -0.002 -0.002
between work & home [0.002] [0.004] [0.003] [0.033] [0.003] [0.002] [0.021]
Observations 320105 230630 189644 151398 297840 282961 267018
Fixed-effects Ind Ind Ind Ind Ind-hw Ind-ww Ind-hw-ww
Stable home ward N Y N Y N N N
Stable work ward N N Y Y N N N
All controls in dummies Y Y Y Y N N N
All controls in trends N N N N Y Y Y
Notes: DEPVAR stands for dependent variable. Clustered standard errors shown in brackets. Distance band 30 kms (excluding ex-
treme wards). Personal and job characteristics include log of firm size (employment) and female, initial 1-dig occupation, initial
age group, type of plant, collective agreement and full-time dummies interacted with a linear trend. Ward 2001 attributes include
mean age of the population, proportion of household living in social housing, proportion of workforce with higher education, un-
employment rate, address density, average distance traveled to place of work and proportion of employees going to work by motor
vehicles. All specification include sector-year dummies (10 categories), commute ring, scheme, distance to scheme and opened
scheme dummies or trends. “Ind” stands for individual, “hw” stands for home ward, “ww”stands for work ward. S.e. clustering
(individual/ward levels). * p<0.1, ** p<0.05, *** p<0.01. Source: ONS, DfT, CASWEB and author’s own calculations.

Table 4: Mobility restrictions and ward fixed effects, log of basic weekly pay

29
D EPVAR : Log of basic weekly pay
1 2 3 4 5 6 7 8
Log of accessibility 0.309*** 0.310*** 0.280*** 0.332*** 0.187* 0.299***
from work ward [0.096] [0.099] [0.099] [0.088] [0.098] [0.093]
Log of accessibility -0.058 -0.057 0.067 -0.042 -0.075 -0.050
from home ward [0.080] [0.084] [0.079] [0.081] [0.079] [0.075]
Log of travel time -0.002 -0.019 0.002 -0.012 0.011 -0.006 -0.001
between work & home [0.021] [0.028] [0.023] [0.025] [0.026] [0.025] [0.021]
Observations 267018 267018 267018 267018 267018 235087 268758 267018
No travel Only travel Only work Only home Excl. same Lagged Instrumental
Robustness Main
time time access access ward access variables
All controls Y Y Y Y Y Y Y Y
Notes: Clustered standard errors in brackets, 2-way clustering work and home ward. Includes all controls. Distance band 30 kms (excluding
extremes). F-stat first stage in column 8 is equal to 1,015. * p<0.1, ** p<0.05, *** p<0.01. Source: ONS, DfT, CASWEB and author’s own
calculations.

Table 5: Robustness: specification changes, log of basic weekly pay

D EPVAR : Log of basic weekly pay

Indiv-ward FX Indiv-house-plant FX
1 2 3 4
Log of accessibility 0.309*** 0.250** 0.264*** 0.275***
from work ward [0.096] [0.100] [0.099] [0.099]
Log of accessibility -0.058 0.022 -0.022 -0.018
from home ward [0.080] [0.083] [0.080] [0.082]
Log of travel time -0.002 0.008 0.003 0.002
between work & home [0.021] [0.019] [0.021] [0.019]
Observations 267018 241624 261007 246818
Fixed-effects Indiv-hw-ww Indiv-hp-pl Indiv-hp-ww Indiv-hw-pl
Always same hward N N N N
Always same wward N N N N
All controls Y Y Y Y
Notes: Clustered standard errors in brackets. “Ind” stands for individual, “hw” stands for home ward,
“ww”stands for work ward, “hp” stands for home postcode, “pl” stands for plant (work postcode-firm-
sector combination). S.e. clustering (individual/ward levels). * p<0.1, ** p<0.05, *** p<0.01. Source:
ONS, DfT, CASWEB and author’s own calculations.

Table 6: Robustness: additional mobility restrictions and FX, log of basic weekly pay

D EPVAR : Log of basic weekly pay

1 2 3 4 5 6 7 8 9 10
Log of accessibility 0.309*** 0.314*** 0.343*** 0.338*** 0.236* 0.349*** 0.354*** 0.392*** 0.396*** 0.263**
from work ward [0.096] [0.099] [0.100] [0.117] [0.121] [0.102] [0.101] [0.099] [0.116] [0.125]
Log of accessibility -0.058 -0.059 -0.118 -0.084 -0.023 -0.081 -0.078 -0.138 -0.109 -0.029
from home ward [0.080] [0.087] [0.097] [0.097] [0.136] [0.093] [0.094] [0.103] [0.103] [0.139]
Log of travel time -0.002 -0.003 -0.011 -0.006 -0.017 -0.003 -0.003 -0.012 -0.008 -0.019
between work & home [0.021] [0.022] [0.022] [0.025] [0.029] [0.023] [0.023] [0.022] [0.026] [0.029]
Observations 267018 253154 236864 218972 187466 230633 225837 215579 202248 176158
Work-home distance All 30 kms 20 kms 15 kms 10 kms All 30 kms 20 kms 15 kms 10 kms
Same scheme N N N N N Y Y Y Y Y
All controls Y Y Y Y Y Y Y Y Y Y
Notes: Clustered standard errors in brackets, 2-way clustering work and home ward. Includes all controls. Distance band 30 kms
(excluding extremes). * p<0.1, ** p<0.05, *** p<0.01. Source: ONS, DfT, CASWEB and author’s own calculations.

Table 7: Robustness: work-home distance restrictions, log of basic weekly pay

30
D EPVAR : Other labour market outcomes
1 2 3 4 5 6 7
Log of accessibility 0.309*** 0.267*** 0.237** 0.196*** 0.199** 0.113* 0.067
from work ward [0.096] [0.097] [0.093] [0.063] [0.078] [0.068] [0.056]
Log of accessibility -0.058 -0.04 -0.044 -0.021 -0.022 -0.037 -0.015
from home ward [0.080] [0.068] [0.118] [0.052] [0.050] [0.063] [0.055]
Log of travel time -0.002 -0.011 -0.071** -0.025 -0.032 0.023 0.021
between work & home [0.021] [0.022] [0.030] [0.023] [0.025] [0.017] [0.019]
Observations 267018 264543 266795 267018 267018 267018 266795
Type Basic Gross Gross Basic Total Basic Total
Outcome Pay Pay Pay Hours Hours Earnings Earnings
Period Weekly Weekly Annual Weekly Weekly Hourly Hourly
All controls Y Y Y Y Y Y Y
Notes: Clustered standard errors in brackets, 2-way clustering work and home ward. Includes all controls. Distance
band 30 kms (excluding extremes). * p<0.1, ** p<0.05, *** p<0.01. Source: ONS, DfT, CASWEB and author’s own
calculations.

Table 8: Other labour market outcomes: wages and hours

D EPVAR : Labour supply by gender

1 2 3 4 5
Log of accessibility 0.180** 0.109 0.297** 0.037 0.400***
from work ward [0.071] [0.068] [0.143] [0.065] [0.099]
Log of accessibility -0.163** -0.062 -0.277 0.057 -0.135
from home ward [0.077] [0.053] [0.180] [0.038] [0.118]
Log of travel time -0.079** -0.027 -0.108** -0.007 -0.038
between work & home [0.031] [0.021] [0.048] [0.019] [0.039]
Observations 267018 131163 135855 131163 135855
Type Full time Full time Full time Weekly Weekly
Outcome Status Status Status Hours Hours
Group All Males Females Males Females
All controls Y Y Y Y Y
Notes: Clustered standard errors in brackets, 2-way clustering work and home ward. Includes all
controls. Distance band 30 kms (excluding extremes). * p<0.1, ** p<0.05, *** p<0.01. Source: ONS,
DfT, CASWEB and author’s own calculations.

Table 9: Labour supply: gender differences

31
References
Ahlfeldt, G. (2011). If We Build, Will They Pay? Predicting Property Price Effects of Transport Innov-
ations. SERC Discussion Papers, 0075.

Ahlfeldt, G. & Wendland, N. (2011). Fifty Years of Urban Accessibility: the Impact of the Urban
Railway Network on the Land Gradient in Berlin 1890-1936. Regional Science and Urban Economics,
41(2), 77–88.

Amiti, M. & Cameron, L. (2007). Economic Geography and Wages. The Review of Economics and
Statistics, 89(1), 15–29.

Baum-Snow, N. (2007a). Did Highways Cause Suburbanization? The Quarterly Journal of Economics,
122(2), 775–805.

Baum-Snow, N. (2007b). Suburbanization and transportation in the monocentric model. Journal of

Urban Economics, 62(3), 405–423.

Baum-Snow, N. (2010). Changes in Transportation Infrastructure and Commuting Patterns in US

Metropolitan Areas, 1960-2000. American Economic Review, 100(2), 378–82.

Baum-Snow, N. & Pavan, R. (2012). Understanding the city size wage gap. Review of Economic Studies,
79(1), 88–127.

Black, D. A., Kolesnikova, N., & Taylor, L. J. (2013). Why do so few women work in New York
(and so many in Minneapolis)? Labor supply of married women across US cities. Journal of Urban
Economics.

Chandra, A. & Thompson, E. (2000). Does public infrastructure affect economic activity?: Evidence
from the rural interstate highway system. Regional Science and Urban Economics, 30(4), 457–490.

Ciccone, A. & Hall, R. E. (1996). Productivity and the density of economic activity. American Economic
Review, 86(1), 54–70.

Combes, P.-P., Duranton, G., & Gobillon, L. (2008). Spatial wage disparities: Sorting matters! Journal
of Urban Economics, 63(2), 723–742.

Combes, P.-P., Duranton, G., & Gobillon, L. (2011). The identification of agglomeration economies.
Journal of Economic Geography, 11(2), 253–266.

Combes, P.-P. & Lafourcade, M. (2001). Transport Cost Decline and Regional Inequalities: Evidence from
France. CEPR Discussion Papers 2894, Centre for Economic and Policy Research.

De la Roca, J. & Puga, D. (2012). Learning by working in big cities. Working paper.

Department of Transport (1998a). A new Deal for Transport: Better for Everyone. White paper - Depart-
ment of Transport.

Department of Transport (1998b). A New Deal for Trunk Roads in England: Understanding the New
Approach. Report - Department of Transport.

Détang-Dessendre, C. & Gaigné, C. (2009). Unemployment duration, city size, and the tightness of
the labor market. Regional Science and Urban Economics, 39(3), 266–276.

Donaldson, D. (2013). Railroads of the Raj: Estimating the Impact of Transportation Infrastructure.
American Economic Review, forthcoming.

Donaldson, D. & Hornbeck, R. (2013). Railroads and American Economic Growth: A Market Access Ap-
proach. NBER Working Papers 19213, National Bureau of Economic Research, Inc.

Duranton, G., Morrow, P., & Turner, M. A. (2013). Roads and trade: Evidence from the US. Technical
report. Mimeo, Wharton School, University of Pennsylvania.

32
Duranton, G. & Puga, D. (2004). Micro-foundations of urban agglomeration economies. In J. V.
Henderson & J. F. Thisse (Eds.), Handbook of Regional and Urban Economics, volume 4 of Handbook of
Regional and Urban Economics chapter 48, (pp. 2063–2117). Elsevier.

Duranton, G. & Turner, M. A. (2011). The fundamental law of road congestion: Evidence from US
cities. American Economic Review, 101(6), 2616–52.

Duranton, G. & Turner, M. A. (2012). Urban growth and transportation. Review of Economic Studies,
79(4), 1407–1440.

El-Geneidy, A. M. & Levinson, D. M. (2006). Access to destinations: Development of accessibility meas-

ures. Report, Research Services Section. Mn/DOT 2006-16, University of Minnesota. Dept. of Civil
Engineering and Minnesota. Dept. of Transportation.

Faber, B. (2012). Trade Integration, Market Size, and Industrialization: Evidence from China’s Na-
tional Trunk Highway System. London School of Economics, mimeo.

Fernald, J. G. (1999). Roads to prosperity? Assessing the link between public capital and productivity.
American Economic Review, 89(3), 619–638.

Fingleton, B. (2006). The new economic geography versus urban economics: an evaluation using
local wage rates in great britain. Oxford Economic Papers, 58(3), 501–530.

Gibbons, S., Lyytikäinen, T., Overman, H., & Sanchis-Guarner, R. (2012). New road infrastructure:
the effects on firms. SERC Discussion Papers, (0117).

Gibbons, S. & Machin, S. (2005). Valuing rail access using transport innovations. Journal of Urban
Economics, 57(1), 148–169.

Gibbons, S. & Machin, S. (2006). Transport and Labour Market Linkages: empirical evidence, implications for
policy and scope for further research. Background paper for the Eddington report to the Department
of Transport.

Gibbons, S., Overman, H. G., & Pelkonen, P. (2010). Wage Disparities in Britain: People or Place?
SERC Discussion Papers, (0060).

Glaeser, E. L. & Kohlhase, J. (2003). Cities, regions and the decline of transport costs. Papers in Regional
Science, 83(1), 197–228.

Graham, D. J., Gibbons, S., & Martin, R. (2009). Transport investments and the distance decay of agglom-
eration benefits. Draft – Department of Transport.

Graham, D. J. & Melo, P. C. (2009). Agglomeration economies and labour productivity: Evidence
from longitudinal worker data for gbs travel-to-work areas. SERC Discussion Papers, (0031).

Gutiérrez-i-Puigarnau, E. & van Ommeren, J. N. (2010). Labour supply and commuting. Journal of
Urban Economics, 68(1), 82–89.

Harmon, N. A. (February 2013). Are Workers Better Matched in Large Labor Markets? Job market paper,
Princeton University.

Harris, C. D. (1954). The Market as a Factor in the Localization of Industry in the United States.
Annals of the association of American geographers, 44(4), 315–348.

Highways Agency (2009). The Highways Agency: Framework Document. Highways Agency - Depart-
ment of Transport.

Holl, A. (2012). Market potential and firm-level productivity in Spain. Journal of Economic Geography,
Forthcoming.

Hymel, K. (2009). Does traffic congestion reduce employment growth? Journal of Urban Economics,
65(2), 127–135.

33
Inhanfeldt, K. R. (2006). A primer on spatial mismatch within urban labor markets. In A compan-
ion to urban economics, Blackwell companions to contemporary economics (pp. 404 – 417). Wiley-
Blackwell.

Jiwattanakulpaisarn, P., Noland, R. B., Graham, D. J., & Polak, J. W. (2009). Highway infrastructure
and state-level employment: A causal spatial analysis. Papers in Regional Science, 88(1), 133–159.

Jofré-Monseny, J. (2009). The scope of agglomeration economies: Evidence from catalonia. Papers in
Regional Science, 88(3), 575–590.

Krugman, P. & Venables, A. J. (1995). Globalization and the inequality of nations. The Quarterly
Journal of Economics, (pp. 857–880).

Krugman, P. R. (1991). Increasing returns and economic geography. Journal of Political Economy, 99(3),
483–99.

Manning, A. (2003). The real thin theory: monopsony in modern labour markets. Labour Economics,
10(2), 105–131.

Manning, A. (2006). A generalised model of monopsony. Economic Journal, 116(508), 84–100.

Manning, A. & Petrongolo, B. (2011). How local are labor markets? Evidence from a spatial job search model.
CEPR Discussion Papers 8686, Centre for Economic and Policy Research.

Marshall, A. (1890). Principles of Economics. London: Macmillan.

Melo, P. C., Graham, D. J., & Noland, R. B. (2010). Impact of transport infrastructure on firm forma-
tion: Evidence from the portuguese municipalities. Transportation Research Record, 2163, 133–143.

Michaels, G. (2008). The effect of trade on the demand for skill: Evidence from the interstate highway
system. The Review of Economics and Statistics, 90(4), 683–701.

Mion, G. & Naticchioni, P. (2005). Urbanization Externalities, Market Potential and Spatial Sorting of Skills
and Firms. CEPR Discussion Papers 5172, Centre for Economic and Policy Research.

Mion, G. & Naticchioni, P. (2009). The spatial sorting and matching of skills and firms. Canadian
Journal of Economics, 42(1), 28–55.

Mulalic, I., van Ommeren, J. N., & Pilegaard, N. (2010). Wages and commuting: Quasi-natural exper-
iments’ evidence from firms that relocate. Tinbergen Institute Discussion Papers, (10-093/3).

Phillips, D. (2012). Getting to Work: Experimental Evidence on Job Search and Transportation Costs. Tech-
nical report. TECHREPORT, Department of Economics, Hope College MI.

Phimister, E. (2005). Urban effects on participation and wages: Are there gender differences? Journal
of Urban Economics, 58(3), 513–536.

Puga, D. (2002). European regional policies in light of recent location theories. Journal of Economic
Geography, 2(4), 373–406.

Redding, S. J. & Venables, A. J. (2004). Economic geography and international inequality. Journal of
International Economics, 62(1), 53–82.

Rosenthal, S. S. & Strange, W. C. (2003). Geography, industrial organization, and agglomeration. The
Review of Economics and Statistics, 85(2), 377–393.

Rosenthal, S. S. & Strange, W. C. (2008). Agglomeration and hours worked. The Review of Economics
and Statistics, 90(1), 105–118.

Rosenthal, S. S. & Strange, W. C. (2012). Female Entrepreneurship, Agglomeration, and a New Spatial
Mismatch. The Review of Economics and Statistics, 94(3), 764–788.

34
Sanchis-Guarner, R. (2012). Essays on Urban and Spatial Economics. PhD Theses. The London School
of Economics and Political Science.

van Ommeren, J. N. & Gutiérrez-i-Puigarnau, E. (2011). Are workers with a long commute less
productive? An empirical analysis of absenteeism. Regional Science and Urban Economics, 41(1), 1–8.

Venables, A. J. (2007). Evaluating urban transport improvements: Cost-benefit analysis in the pres-
ence of agglomeration and income taxation. Journal of Transport Economics and Policy, 41(2), 173–188.

Vickerman, R. (2002). Transport and Economic Development. Transport and economic development
round table 119, European Conference of Ministers of Transport, OECD.

Vickerman, R., Spiekermann, K., & Wegener, M. (1999). Accessibility and economic development in
europe. Regional Studies, 33(1), 1–15.

Zenou, Y. (2009). Urban Labor Economics. Cambridge University Press.

35