The name is absent

implicit insurance, more contractual insurance is bought than implicitly generated through rents, c* < b*.

With brokers, the “renewal rights” to a book of business are transferred to the broker under its brokerage
contract with the insurer. This magnifies the hold-up which results in larger negotiated transfers of non
verifiable (but ex post insurable) losses as shown by b*r. Moreover, the interest domain over which transfers
are made, is expanded from r to rb_r. Not only do brokers enable more complete transfers of the non
verifiable loss, they permit an efficiency gain though diversification. This arises because the brokers can
use the future rents from policyholders not having a loss, to expand its hold-up on behalf of those who do
suffer a loss. The increased hold-up power reduces the implicit premium loading and thereby increases the
transfers for non-verifiable losses, b*_r > b*. This, in turn, reduces the negative spill-over effect — the spread
in marginal utilities is lower — and increases the transfers for verifiable losses, c*_r > c*. Thus, with brokers,
more contractual insurance for verifiable losses is bought and this is accomplished with lower transaction
costs imposed on the extra-contractual market for non-verifiable losses.

The role defined here for brokers raises some topical challenges. Brokers have recently been under a
well publicized attack from the attorney general of New York, Elliot Spitzer. This attack has decimated
the value of the world’s largest broker and largest insurer, forced the resignation of both their CEO’s, led
to several criminal indictments, and has resulted in all four of the largest brokers abandoning a major
form of compensation. At the heart of the issue is the compensation structure for brokers. In addition
to commissions (proportional to premiums), brokers often receive “contingent commissions”. These are
payments by the insurer to the broker based on some measures of the book of business transacted. Most
common are contingent commissions based on the profit of the broker’s book, but volume based contingent
commissions and renewal based structures also are found. The allegation of Spitzer, is as follows. The
broker is an agent of the policyholder. Therefore, broker compensation linked to the welfare of the insurer
must compromise its obligations to its principle (the policyholder).

Our results challenge this simplified conclusion. We have shown that brokers can benefit policyholders
by expanding the market for non verifiable losses. To do so, they must exercise a credible threat to
sanction insurers that do not respond to such losses. This is achieved by constructing a similar incentive
compatibility constraint for the insurer and broker such that they both gain from the ex post transfer, i.e.
that the broker shares in the future rents from the book it holds with the insurer. In our formal model,
the broker compensation is linear in the rents. This contradicts Spitzer’s blanket assertion that “kickbacks”

18

<<   17   18   19   20   21   22   23   >>

More intriguing information