Very often, the social cost
from corruption
isn't really well captured by the notion
of people having to pay bribes.
The true and proper social costs have
to do with lower quality social outcomes.
We're going to look
at the country of India
and the topic of corruption
in the issuing of driver's licenses.
I'll be drawing on a very good paper
on corruption and it is entitled
"Obtaining a Driver's License in India".
The author set out to answer
a pretty simple question:
Should this guy be on the road or not?
It turns out the answer
is going to be a little scary
and that's because of corruption.
The author set up an experiment
to try to figure out
how corrupt was the process
of issuing driver's licenses in India.
There are three different groups involved.
First, they took a randomly selected group
and offered them a bonus if
those individuals could get a license,
especially fast.
The quickest you could get a license in
this context was within 32 days.
There is a second group,
also randomly selected
and that group was taken aside
and given driving lessons
There was a third group
and that group was left alone
as they just navigated the process
of getting a driver's license in India.
At the end of all of this investigation,
what the authors are going to do
is compare the three groups
and see what we can learn
from that comparison.
The important point here is that
one group has a special incentive
to navigate the process
especially quickly,
that's the first group.
The people in that first group;
they generally hired
a bunch of individuals
who are called agents.
These agents help you navigate
the driver's license process in India.
The agents, in effect,
they bribe the process for you.
You don't have to engage
in the corruption yourself.
You hire the agent to help you.
The agent knows the ins and outs
of the bureaucracy.
And, basically, the agent
issues some bribes
to make sure you can get a license,
and you can get a license quickly.
That's what's going on
with the first group.
So, what are the results?
This first group of individuals,
who are paid the bonus
if they got licenses quickly;
they did in fact get the licenses quickly
due to the aid of the agents,
and they got the licenses
at a higher rate.
71% of the group paid the bonus,
got the licenses,
whereas, in the control group,
only 48% of them
managed to get the licenses.
In other words, some of the individuals
were paying to get the licenses,
when really maybe they should not
have had the licenses at all.
You can think of this picture
as one instance of
the social costs of corruption.
Just how big is this problem?
Well, it turns out
it's a somewhat grim story.
Of the bonus group,
only 38% had to actually take
and pass a driving test to get a license
That's because of the aid
of these corrupt agents.
Those who were in the bonus group
who got a license,
they were then independently tested
to see whether or not
they were good drivers.
It turns out that 65%
of them failed that test,
and maybe those individuals
should not be on the road.
This is, again, another lesson in
the high social costs of corruption.