Nyberg: How do conflicts of interest in universal banking emerge and what are the arguments for a separation of commercial and investment banking?
I would like to begin by thanking the Swedish Bankers’ Association, the Swedish Securities Dealers Association and Finansinspektionen (the Swedish financial supervisory authority) for taking the initiative to organise this seminar. Over the past year, in particular, we have seen the importance of open and thorough discussion of conflicts of interest in the financial sector. I hope that this seminar will contribute to an even deeper analysis of the problems we face.
The present discussion has arisen in the wake of a number of international corporate scandals. It is perhaps sufficient to mention names such as Enron, WorldCom and Skandia to indicate how much trouble can be caused by conflicts of interest.
In Sweden there are two conflicts of interest in particular that have received attention, although there are many other, possibly equally serious ones. The first is that many sellers of various financial services offered by banks and insurance companies have traditionally been presented as investment advisors rather than salespeople. The risk of a conflict of interest lies in the fact that their suggestions for investment have been designed more to promote the seller’s or the company’s interests rather than to fulfil the needs of the investor. The second example comes from the life insurance sector. When the life insurance part of a company buys services from the property and liability insurance part, there may be incentives to deviate from correct internal pricing, especially when one part of the company is a mutual company and the other is a profit-making company. Another problem in an insurance company is how a surplus – or a deficit – should be distributed between different groups of insured parties. The question is whether, for instance, those receiving pensions now should share the burden of the insurance companies' poorer income, or whether they have had the level of their pension payments guaranteed.
However, today we have gathered to discuss conflicts of interest in banks, and I have promised to talk about the classical conflict between commercial banking and investment banking. The conflict lies within the universal bank’s double role as lender and investment adviser. If a bank’s borrowers suffer deterioration in their financial prospects, the bank can reduce its credit risk by helping with new share issues, etc. This conflict of interest has been discussed in the United States for several decades now, but in Sweden it has scarcely been mentioned. There is reason to ask why the differences are so great, and I shall return to this point.
Another example, which has recently led to new legislation in the United States, is the role of analysts in investment banking. An analyst has the task of providing investors with advice, but if the bank is more interested in gaining business for its corporate finance activities than safeguarding the investor's interests, the analyst’s assessment may be influenced by this. Here there should traditionally be a “Chinese wall”, but this has in some cases been fairly low and probably eroded considerably over time.
Before I go on to more bank-specific issues, I would like to say a few words about conflicts of interest in general, and how these can be handled – to the extent that they should be handled. I shall then move on to more specific conflicts of interest between different types of banking operations, particularly given the developments in the United States. Finally, I shall draw some conclusions concerning Sweden.
The fact that different people have different interests is quite natural. A buyer and a seller usually have different interests in a business transaction. A difference of interest is actually one of the most important driving forces in a market economy. A difference of interest only becomes a conflict of interest when one person or organisation has to take into account several different interests and can promote one at the cost of another. Conflicts of interest are thus based on an agency relationship. These conflicts exist in almost all businesses. Employees shall usually work in the best interests of the company, but can in some contexts have interests of their own that are in conflict with the company’s best. Divisional managers within a corporate group may have an interest in promoting their own division rather than the group as a whole. Company management does not always have the same aims as the shareholders. Many of these conflicts of interest are common and quite natural elements of real life, and rarely cause problems. It is important to bear this in mind when discussing methods of dealing with them.
This does not mean that conflicts of interest in financial operations are irrelevant, quite the reverse. The relevant question here, however, is which conflicts of interest are real problems and which are not.
The conflicts without problems are those where there is little incentive to improperly favour one interest over another. For instance, most employees are usually loyal to their employer – otherwise they risk losing their jobs. A company that does not take into account the interests of its employees and its customers risks gaining a bad reputation and losing its competitiveness.
Other conflicts of interest are more problematic. When an agent has a considerable incentive to favour one party at the cost of another, there could be scope for some form of regulations. Many companies also have ethical guidelines and rules for dealing with the cases where there is most incentive to take other interests into account. The discussion on corporate governance is one example of how the market is trying to find solutions to various conflicts of interest. It is not usually necessary to have any regulation by the authorities in this area. Usually it is sufficient to have contracts between the parties or to create norm-based standards within the sector.
However, there is a further dimension to the conflicts of interest that concerns the interests of society. Promoting one party at the cost of another usually entails a redistribution of economic resources. This may be morally dubious without always involving large costs to society. There is a bigger problem when conflicts of interest hide or distort information. This leads to real economy decisions being made on the basis of inferior or incorrect information. The recent scandals are serious because the people who were supposed to provide investors with reliable information had incentive to conceal this information in order to promote their own interests, or those of their company.
It is the role of the financial sector as distributor of information in the economy that makes financial conflicts of interest particularly interesting. People in the financial sector are responsible for assessing which companies are robust and which are therefore granted access to growth capital. If conflicts of interest within the financial companies disturb this control function, it may lead to substantial costs to society.
Which measures need to be taken to ensure that conflicts of interest do not lead to financial agents concealing, distorting or incorrectly using information? There are some different potential solutions. I have already observed that the market itself is often capable of solving the problems. This can be achieved, for instance, by customers making demands for information on various conflicts of interest. However, the collection and distribution of information is not free. And as information contains a strain of public goods, there is a risk that the market solution will generate too little information. It may therefore be necessary in certain contexts to have regulations stipulating that information on conflicts of interest must be made public. Regulations concerning greater insight into various persons’ incentives may limit the effects of different conflicts of interest. It would help outsiders to assess the quality of the information provided. A more drastic solution would be to require that activities exposed to conflicts of interest should be managed by different companies, or even different groups. The Glass Steagall Act in the United States, for instance, required that investment banks and commercial banks must be legally separated.
However, there are clear disadvantages with forcing a separation of different activities through legislation. All regulation also entails costs – both explicit and implicit costs. It is important from society’s point of view that these costs do not exceed the usefulness of the regulation. One cost of separating various activities is that it makes it difficult, or impossible, for financial intermediaries to benefit from synergies and economies of scale.
A clear example of this is contained in the conflict between traditional commercial banking and investment banking, which is my subject today. A bank that lends money to a company normally gains access during the credit application process to special information about the company – information that is not publicly available. The bank makes an investment in the information on the company and therefore usually has an information advantage in relation to other banks and outside parties. This investment can also mean that the bank can offer the company investment banking services in a more efficient manner. The fact that a company seeking funding in the capital market has access to bank credit is often regarded by the market as a positive signal. It means that a professional agent has analysed the company and found it creditworthy, i.e. been willing to risk his or her own capital in it. The information about the credit functions is a quality signal or a certification.
At the same time, opportunities arise to make use of conflicts of interest when a bank functions as both credit-granting institution and adviser or guarantor in new issues. If the prospects for a company that has loans with the bank deteriorate, it may be in the interests of the bank to reduce some of its credit risk by aiding the company with a new share issue. In this way, the bank can pass on some of the credit risk to the shareholders. From society’s point of view, there must be a balance between potential efficiency gains and risks of exploitation in conflicts of interest. Whether or not legislation is required depends on the seriousness of the conflict of interest and the size of the incentives to promote one's own interests.
Examples in the United States
The clash between utilising economies of scale and limiting the damaging effect of conflicts of interest goes back a long way, particularly in the United States. Following the stock market turbulence in 1929, there was intensive discussion of these conflicts of interest. Now they have come into focus once again, following a new stock market crisis. It appears as though fewer people care about the conflicts of interest when earnings are high.
The US 1930s example began back in the mid-1920s, when stock market prices soared. Many people who bought shares during this period were convinced that this long upturn meant a new epoch with permanently higher earnings on shares. The insight that prices could not continue to rise at the same rate spread gradually and the stock market fell heavily in autumn 1929. The fall continued for several years. After three years, the value was down to 15 per cent of the previous figure. Many investors felt that they had been deceived by the banks who had sold them shares and bonds. In several cases it was the new universal banks that had been most active in mediating securities. These universal banks combined traditional bank lending with various forms of investment banking activities. By using economies of scale from these different areas, they were able to offer competitive prices – but also to exploit various conflicts of interest.
When prices fell, the banks were accused of having contributed to issues of securities of dubious quality and of having painted an overly bright picture of the companies’ future in their prospectuses. The commercial banks were also accused of having converted loans lacking creditworthiness into shares, by contributing to new issues, where shares were sold to an uninformed public. In addition, it was claimed that certain company management had lined their own, and their families’, pockets in various ways.
Although it was difficult to prove that the banks had knowingly and systematically misled the general public, it was revealed that prominent figures in various banks had earned large sums at the cost of their customers, and in some cases at the cost of the banks. This contributed to the introduction of the Glass Steagall Act in the United States in 1933. The Act required complete legal separation of commercial banks and investment banks. A commercial bank could not be included in the same group as a bank that mainly conducted operations in the securities markets.
However, during the 1960s and the 1980s the regulations in the United States were gradually loosened up. Commercial banks were given increasing opportunities to conduct investment banking activities, albeit on a limited scale. During the 1990s, a debate arose in the United States as to how far these exceptions could be made and whether the universal bank was to be or not to be. A number of scientific studies showed that in the trade-off that existed between benefiting from economies of scale within different types of banking activities and increasing the risks of exploiting conflicts of interest, the benefits of economies of scale weighed heaviest. For instance, it was shown that the shares and corporate bonds issued between January 1927 and September 1929 gave lower earnings if they were issued by a commercial bank than if they were issued by an investment bank. This indicated that investors required a lower risk premium if a commercial bank managed the issue than if an investment bank was responsible. The investors appear to have trusted commercial banks more, and fear of exploitation of the conflicts of interest does not appear to have dominated. The investors appear to have had most confidence in the issue when the commercial bank had outsourced its investment banking activities in a separate company. This would indicate that the investors were aware of the conflicts of interest, but that they considered the commercial banks' greater capacity for certification outweighed these. In 1999 these regulations were completely removed by the introduction of the Gramm-Leach-Bliley Act.
After the fall in share prices we have seen over the past four years, the focus of the general debate has once again landed on the conflicts of interest that exist between the different areas of a bank's operations. The repealing of all parts of the Glass Steagall Act is now being questioned, and there have once again been demands for different types of regulation.
What bearing does this have on Sweden?
The question is what we in Sweden can learn from the American example. In my opinion, it shows that there are potentially dangerous conflicts of interest between traditional commercial banking and investment banking. The question is whether we need to do anything about this in Sweden. Can these conflicts be handled by means of the markets, by sector norms and standards, or is some form of public regulation required? I draw four conclusions here.
Let me take the most important one first. I do not believe that it is appropriate to use regulation to forbid dealings between traditional commercial banks and investment banks. Nor do I believe that it is necessary at present to introduce regulation that would force different legal entities to be formed for investment banks and traditional commercial banks. They should be able to remain departments in the same legal entity. There are two arguments in favour of this. The first argument against regulating to separate banking activities is that it could limit the positive economies of scale and synergies. The second argument is that there is no evidence that conflicts of interest between the parts of our Swedish banks conducting investment banking activities and the traditional lending activities have led to abuse.
One can, of course, speculate as to why abuse of this type of conflict of interest does not occur in Sweden in the same way as in the United States. I have no clear answer to that. However, there are a couple of institutional differences. Share issues in Sweden have rarely been guaranteed or syndicated in the same way as they are in the United States. The role of the Swedish investment banks in share issues has therefore often been less comprehensive than that of their US counterparts. This may have limited the problem.
Another institutional difference is that the bank market is more concentrated in Sweden. This may make the banks more anxious of their credibility than US banks are. If a Swedish bank were to pass on its own credit risks to investors in a share issue, there is a greater risk that this would seriously damage the bank’s credibility. The number of large Swedish investors is fairly small and it would quickly become known if anyone felt they had been cheated. The incentives to exploit a conflict of interest have quite simply been much lower here than in the United States. This does not necessarily mean that Swedish bankers are more honest than American ones, just that the advantages of exploiting conflicts of interest are less and the costs are greater.
My second conclusion is that it is important that all significant conflicts of interest should be discussed openly and publicly. There are justifiable claims for reliable information on borrowers and issuers, but the same applies to information on the measures taken by the banks to avoid exploiting conflicts of interest. I believe that there are relevant market solutions in most cases, as long as the conflicts of interest are clear.
Thirdly, I would like to encourage the Swedish banks to carefully follow international developments within the field of compliance – i.e. the routines to ensure that internal and external rules are adhered to. In many cases the requirements have become much more stringent in recent years, and it is important to thoroughly assess developments in other countries. For instance, the Chinese wall has been repaired in the United States, and is now higher than ever. This has been achieved through legislation. It remains to be seen whether the Swedish banks can solve these problems themselves.
Fourthly, I believe that the supervisory authorities play an important role with regard to identifying and overseeing potential conflicts of interest. Our Finansinspektion has begun to do this to a greater extent, not only with regard to banks but for the financial sector as a whole. Although the market itself can manage most of the conflicts of interest, it may take some time before a common practice is established. Experience also shows that good practice and good ethics can be eroded rather quickly in certain market situations. This is where the supervisory authority can help to develop and maintain confidence in the financial sector as a whole.
Thank you.
