Centre for Economic Policy Research

General discussion
Jean-Pierre Landau said that, before thinking of remedies, one must first ask why
conflicts of interest arise in the first place. The Report takes the view that they
develop because some firms undertake multiple functions. So, when looking at
the remedies, there is a trade-off to be considered between the economies of scope
generated by such a bundling of activities and the risk of conflicts of interest.
Another explanation would see conflicts of interest as a by-product of the 
difficulty – or impossibility – of charging a proper price for services such as 
ratings, analysts or audit. In an efficient market one should not have to pay for 
information since it is already in the price. It is costly, however, to collect and
process information, so the question arises as to how it gets into the price in the
first place. This is a well-known paradox about market efficiency. So we could ask
whether conflicts of interest result from attempts to circumvent this paradox, by
getting investors to pay indirectly for information. This can be done in many
ways. The first one is cross-subsidizing activities so that information can be 
produced, although people will not buy it for its real price. Analysts cannot 
recover their costs, and that is why they have been used for other purposes. This
cross-subsidization is at the source of conflicts of interest. The second way is to
exploit the rents created or consolidated by regulatory privileges. This is 
obviously the case for rating agencies. At issue here is not so much a conflict of
interest but the quality of the information produced in an environment of very
low competition. 
Discussion and Roundtables 103

---

Information is a public good and attempts to treat it as a private good are
bound to create distortions, some of them in the form of conflicts of interest.
Would it not be more appropriate to recognize from the start that we are faced
with a public good problem? The goal, then, is to find the least distortive way of
financing these public goods. Would partial ‘socialization’ be a solution in some
cases? It might not be necessary for audit services. A mix of good governance and
supervision might work. The question of paying for information does not arise in
this case, because companies are legally required to have their accounts audited
and pay for it. As far as rating agencies are concerned there must be a way to
finance them, while at the same time taking the rents out of them. This calls for
encouraging entry in that market and by helping the creation of new agencies,
possibly through subsidization. In the case of analysts, a degree of public 
intervention is provided in the ‘New York’ settlement, by the obligation to buy
research from independent firms. So the obligation to pay for information is made
transparent and explicit. Landau identified this as a progressive move, since it will
create and stimulate the independent research industry. He admitted that he
could not see the drawbacks and questioned whether there is any empirical basis
to the argument that it will lower the quality of information. After all, top 
qualified people – including the present Chairman of the Fed – have made a career
in independent research companies.
Frederic Mishkin agreed with the first part of Landau’s argument, that the
impossibility of charging people for the information which may be used for 
multiple uses is a source of conflicts of interest. This does not mean, however, that
the firm as a whole cannot charge for information. Mishkin disagreed with the
second part of the comment dealing with socialization. Bundling is a way to
charge for information. This is not only true for the financial services industry but
also for many other industries where synergies exist between different activities or
components. For example, cars come with radios and, yet, we think that the 
market works well.
Mishkin also agreed with the view that solving the conflict of interest problem
in the financial services will not solve the key corporate governance problem. He
also agreed strongly with the need for market participants to be involved in the
design of policy. This is exactly the way good supervisory oversight operates: 
public authorities should gather the view of the market participants and 
synthesize them in the policy framework.
Alexander Swoboda said that many recent problems that have received so
much attention in the popular press are not real issues of conflicts of interest;
rather they reflect basis misconduct and outright crime and should be reported as
such. Transparency is necessary, but not sufficient to generate adequate 
information. As shown by the bubble or the international financial crises, there
are periods when the incentives to use the information are downplayed for 
psychological or other reasons. This is where it would be most important to have
good rules in place because the costs of conflicts of interest are hidden during the
upward part and only get revealed after the crash. It remains important to have
codes of conduct and best practices, as they provide the regulator with a 
benchmark on which to judge whether the compensation scheme is adequate. It
is difficult, however, to devise the scheme and also to ensure that it does not
become a hard regulation that is stifling to the development of innovation.
Antonio Borges commented on corporate governance, and pointed out that
there can be a conflict of interest among shareholders. One of the key differences
between the United States and Europe is the way in which companies are 
controlled. Companies within the United States are widely held by numerous
104 Conflicts of Interest in the Financial Services Industry

---

shareholders and the principal-agent problem is acute. Managers sometimes act
against the interests of shareholders, and get away with it. In continental Europe
the vast majority of corporations are controlled by so-called reference 
shareholders, players with very large blocks of shares, often majorities, who have
complete control over management, and who replace the agent at the slightest
hint of a principal-agent problem. That is why many of the problems analysed
here do not occur in Europe. The real problem in Europe is that management and
majority shareholders collude against the minority shareholders. This is a real
conflict of interests. Recent reports have emphasized the role of auditors, of 
powerful independent directors, of a strict regulation to require disclosure 
of transactions between corporations and officers and between corporations and
significant shareholders, and the fact that there would be much less opportunity
for taking advantage of minority shareholders if all of that were publicly known.
Unfortunately this is not happening in continental Europe.
Discussion and Roundtables 105

---

---