 |
|
|
I ran across this in HBR Daily
Stats:
MARCH 22, 2013
Higher Levels of Social Capital Lead to Fewer Traffic
Fatalities
A 5% increase in the average level of agreement with the
statement "Most people are honest" within a U.S. state results
in a decline in traffic fatalities in that state by about 11%,
says Matthew G. Nagler of the City College of New York. This
measure of trust in others is an indicator of the state's
level of social capital, sometimes defined as a willingness to
engage in community activities. Less-conscientious people who
reject civic engagement presumably drive more recklessly.
Source: Does Social Capital Promote Safety on the Roads?
Since this is full of pattern
information, I sought to find out more:
I attached the pdf I found, fyi,
and copied some of here so you don't have to open it, because
this piece is loaded with information at the meta level:
1) Social capital as a
metaconstruct
2) How to skirt previously direct
links looking for indirect causal links (something in this, I
don't fully understand, but I noticed it's there)
3) Recognition that something
besides war kills so many people, and how it's escalating
exponentially
4) If this kind of noise/signal
finding works, then we can expect it to become much more
prevalent...?
|
|
Helpful Hint:
I don't know why HBR chose
something like this, as I don't know their editorial
guidelines...it could be for the same reasons I mentioned I
was attracted to it...it would be interested to see how many
people in their database actually clicked on some part of this
for more research, I suppose (for the curious)...and for the
practical, is "social capital" really an important prime in
correlating cause and effect issues, in something as simple as
traffic fatalities and if so, then what else can we
predict/explain with social capital? |
|
Action Step:
And if so...will social capital
start appearing more widely in our policy making
decisions/design?
In sociology, social capital is
the expected collective or economic benefits derived from the
preferential treatment and cooperation between individuals and
groups. Although different social
sciences emphasize different aspects of social capital,
they tend to share the core idea "that social
networks have value".
Just as a screwdriver (physical
capital) or a university education (cultural
capital or human
capital) can increase productivity (both individual and
collective), so do social contacts affect the productivity of
individuals and groups.[1]
I found this expose on the
definition of social capital because I had not seen social
capital used in the manner in which the HBR Stats presented it
before, another reason it caught my mind, not just my eye...
http://www.socialcapitalresearch.com/literature/definition.html
Now, here is the piece on
MATTHEW G. NAGLER The City College of New York, mnagler@....
I present evidence that social capital reduces traffic
accidents and related death and injury, using data from a
ten-year panel of 48 U.S. states. The econometric challenge
is to distinguish the causal effects of social capital from
bias resulting from its correlation with unobservable
characteristics by state that influence road risks. I
accomplish this by employing snow depth as an instrument, and
by restricting attention to summertime accidents. My results
show that social capital has a statistically significant and
sizable negative effect on crashes, traffic fatalities,
serious traffic injuries, and pedestrian fatalities that holds
up across a range of specifications.
JEL Classification: R41, I18, Z13 Keywords: Highway safety;
Panel data; Instrumental variables; Social factors;
Trust
1
I. INTRODUCTION
Social capital has generated substantial academic interest
over the past two decades in the
wake of two influential books by Robert D. Putnam, Making
Democracy Work (1993) and
Bowling Alone (2000). A large number of recent studies suggest
that interpersonal trust and civic
engagement have positive economic impacts (e.g., Narayan and
Pritchett 1999, Knack 2001, Zak
and Knack 2001, Grootaert et al. 2002, Karlan et al. 2009).
Recent work also indicates that social
capital, measured variously, has beneficial impacts on health
and well-being.1 The research
attention has, however, brought with it a measure of
controversy. While there is much evidence
of association between social capital and outcomes relevant to
economics, causal links have been
hard to establish conclusively (Helliwell 2001). The lack of
evidence of causation flowing from
social capital to putative outcomes has been one of the
central critiques of this literature (see,
e.g., Sobel 2002).
This paper examines how social capital relates to highway
safety. It uses an aggregate
measure of generalized interpersonal trust to explain
variations in the level of traffic fatalities
and three other measures of highway safety across a panel of
U.S. states over the years 1997-
2006. The focal point of my approach is an innovative
identification strategy that solves the
problem of establishing causation.
The main difficulty with estimating the effect of social
capital on highway safety at an
aggregate level is that selection on unobserved
characteristics of the population may introduce
spurious effects. For example, if less conscientious
individuals who both eschew civic
engagement and drive more recklessly tend to sort
disproportionately across states, one might
observe higher rates of fatal accident and other highway risk
measures in states with lower levels
of social capital. This, of course, would prove nothing about
the effects of social capital per se.
2
The paper addresses this problem by exploiting variation in
winter snow depth across
states as an exogenous source of variation to social capital
formation. Snow depth offers a
relevant instrument because a snowy climate impacts the
long-term movement patterns of
individuals. These in turn are relevant to the extent to
which individuals form strong ties with
each other. But for this instrument to be valid, it must also
be orthogonal to unobservable
determinants of highway risks. As snow accumulation likely
contributes directly to the
incidence of accidents during the winter, I restrict the
dependent variable to safety-related
incidents occurring during the summer.2 Conceptually, while
variation in snow depth plausibly
explains variations in social capital from state to state, it
does not directly affect the rate of
accidents in the summer (non-snow) months of the year.3 Thus
I am able to investigate how
relative differences in social capital – for which variation
is induced through exogenous snow
depth variation – influence non-snow-related rates of traffic
incidents.
I find that social capital has a significant negative impact
on the incidence of crashes,
injuries and deaths on the roads. For example, a
one-standard-deviation increase in my main
social capital variable – the average individual agreement
level with the statement, “Most people
are honest” – results in a decline in state crashes of between
9 and 18%, a decline in traffic
fatalities of between 11 and 19%, a decline in serious
injuries of between 11 and 22%, and a
decline in pedestrian fatalities of between 20 and 39%.
A number of previous studies have used ordinary least squares
or fixed effects regression
to estimate the economic effects of social capital (e.g.,
Helliwell 1996, 2007; Narayan and
Pritchett 1999; Maluccio et al. 2000; Grootaert et al. 2002;
Yamamura 2008). While some
studies have attempted to account for selection bias by
employing instrumental variables
techniques (e.g., Knack and Keefer 1997, Narayan and Pritchett
1999, Maluccio et al. 2000, Zak
3
and Knack 2001, Easterly et al. 2006), the instruments have
largely consisted of other
sociopolitical variables. Given the nature of relationships
among social science variables,
questions remain about the endogenity of such instruments and
whether they may be reasonably
perceived as meeting the exclusion criteria. With its
“natural” (i.e., climatic) experimental
design, the present study offers an important innovation in
conclusively demonstrating the
effects of social capital.4
An important motivation for this study is that traffic
fatalities pose a serious public health
problem worldwide. Over one million deaths are attributed each
year to automobile accidents
(Bishai et al., 2006). If current trends are not reversed,
traffic injuries are projected to grow from
the ninth leading cause of disability-adjusted life years lost
in 1990 to the third by 2020 (Murray
and Lopez, 1996). In the United States, a trend of declining
traffic deaths during the 1970s and
1980s, attributable to the use of seat belts and adoption of
safety equipment such as antilock
brakes, has yielded to stagnation since the 1990s; the toll in
the U.S. remains near 43,000 deaths
annually (White, 2004). Traffic fatalities remain a major
cause of death at all ages and the
leading cause for persons under the age of 44 (Heron et al.,
2009).5
The rest of the paper is structured as follows. Section II
considers conceptually why
social capital should affect highway safety. Section III lays
out my empirical strategy. Section
IV describes the data used in the study. Section V evaluates
the instrumental variables
mechanism. Section VI presents results and robustness checks,
and Section VII discusses public
policy implications. |
|
Comments: |
|
