IV. Factors Behind Differential - 2

Another common explanation of differential
corruption, popular among sociologists,
is that social norms are very
different in different countries. What is
regarded in one culture as corrupt may
be considered a part of routine transaction
in another. (Visiting Westerners
are often aghast that an Asian or an African
will sometimes not carry out his
ordinary service without baksheesh or
tips; the latter, on the other hand, finds
the high degree of monetization even in
personal transactions in advanced capitalist
countries somehow “corrupt.”)
But a more important issue is involved.
It is widely recognized that in developing
countries gift-exchange is a major
social norm in business transactions,
and allegiance to kinship-based or clanbased
loyalties often takes precedence
over public duties even for salaried
public officials. Under such circumstances
use of public resources to cater
to particularistic loyalties become quite
common and routinely expected. At the
same time, it will be wrong to suggest
that concern about public corruption is
peculiarly Western. In most of the same
developing countries, public opinion
polls indicate that corruption is usually
at the top of the list of problems cited
by respondents. But there is a certain
schizophrenia in this voicing of concern:
the same people who are most vocal
and genuinely worried about widespread
corruption and fraud in the
public arena do not hesitate at all in
abusing public resources when it comes
to helping out people belonging to their
own kinship network. (It is a bit like the

U.S. Congressmen who are usually livid
about the rampant pork-barrel politics
they see all around them but they will
fiercely protect the “pork” they bring to
their own constituency.) Edward C.
Banfield (1958) comments on the prevalence
of what he calls “amoral familism”
in the Mezzogiorno in Italy, but Robert
Putnam (1993) observes in his study of
comparative civicness in the regions of
Italy that the amoral individuals in the
less civic regions clamor most for
sterner law enforcement. Mayfair Yang
(1989) notes how people in China generally
condemn the widespread use of
guanxi (connections) in securing public
resources, but at the same time admire
the ingenuity of individual exploits
among their acquaintances in its use.
A major problem with norm-based explanations
is that they can very easily be
near-tautological (“a country has more
corruption because its norms are more
favorable to corruption”). A more satisfactory
explanation on these lines has to
go into how otherwise similar countries
(or regions in the same country like
North and South in Italy) may settle
with different social norms in equilibrium
in, say, a repeated game framework,
and how a country may sometimes
shift from one equilibrium into
another (as has happened in the case of
today’s developed countries in recent
history with respect to corruption).
The idea of multiple equilibria in the
incidence of corruption is salient in
some of the recent economic theorists’
explanations. The basic idea is that corruption
represents an example of what
are called frequency-dependent equilibria,
and our expected gain from corruption
depends crucially on the number of
other people we expect to be corrupt.
At a very simple level the idea may be
illustrated, as in Andvig (1991), with a

so-called Schelling diagram shown in
Figure 1. The distance between the origin
and any point on the horizontal axis
represents the proportion of a given total
number of officials (or transactions)
that is known to be corrupt, so that the
point of origin is when no one is corrupt,
and the end-point n is when everyone
is corrupt. The curves M and N represent
the marginal benefit for a
corrupt and an honest official respectively
for all different allocations of the
remaining officials in the two categories.
The way the curve N is drawn, the
benefit of an honest official is higher
than that of a corrupt official when very
few officials are corrupt, but it declines
as the proportion of corrupt officials
increases and ultimately becomes even
negative when almost all others are
corrupt. The M curve goes up at the
beginning when more and more officials
are corrupt (for the marginal
corrupt official lower reputation loss
when detected, lower chance of detection,
lower search cost in finding a
briber, etc.), but ultimately declines
(when the size of bribe is bid down by
too many competing bribers, for example),
even though at the end-pont the
pay-off for a corrupt official remains
positive.