Modern Banking

[ 538 ]

M O D E R N B A N K I N G

a net subsidy from a government safety net, was the principal benefit of the mergers, suggesting the regulators were correct to disallow the mergers.

The longer term (i.e. over 10 years) effects of mergers are more difficult to pinpoint but may be more favourable. Davis (2000) notes that the European banks that merged in 1990 suffered a loss of market share, took longer than expected to get the cost/income ratio down (BBV took seven years before the cost to income ratio returned to pre-merger level), and it can take many years to resolve culture clashes. So shareholder value may suffer for several years, but the key question is: what would happen in the absence of the merger? One can never answer this critical counterfactual question definitively. However, it seems likely that many banks that did merge might well have fallen victim to hostile takeovers otherwise, and some, left on their own, could have drifted into insolvency.

9.6. Conclusion

This chapter has explored key competitive issues as they relate to banking markets. Differences in banks prompt a number of questions, the answers to which are important to managers (especially those working in the area of bank strategy) and policy makers. The key questions are:

žHow much, if anything, do financial firms stand to gain by expanding their output and/or product range?

žHow widely do banks vary in the efficiency with which they use resources?

žHow do banks’ average costs vary with their size?

žShould banks specialise or produce multiple outputs?

žIs the banking sector dominated by a few large banks that charge higher/lower prices, and if either, why?

žAre bank mergers and acquisitions (M&As) beneficial for shareholders?

žAre bank M&As good or bad for customers?

žTo what extent does the possibility of entry keep the banks from overcharging?

The empirical evidence on these issues is complex and rarely clear cut. Some questions are hotly disputed. Others elicit answers that vary according to the country or period studied, or the methodology employed. However, it is possible to summarise a number of points from the material reported in this chapter.

First, there is wide diversity in the degree of ‘‘X-efficiency’’ of banks, both within countries and between them. The Berger and Humphrey (1997) survey found that management could improve X-efficiency anywhere between 10% and 30%. It is worth stressing a major drawback with these measures: they are relative, so an entire banking system could be inefficient even though the cost X-efficiency estimates suggest otherwise.

The presence or otherwise of economies of scale is an important issue for bank managers for several reasons. If they are only present for small banks, as much of the literature using data from the 1980s seemed to suggest, then mergers and/or organic growth could make things worse. Data from the 1990s and more sophisticated econometric techniques have

[ 539 ]

C O M P E T I T I V E I S S U E S I N B A N K I N G

produced evidence of scale economies at much larger asset sizes, for both US and European banks. In other words, increasing their scale of operations will reduce average costs. If economies of scale are present, they can be used by incumbent firms as an entry barrier, which could encourage anti-competitive behaviour, possibly prompting intervention by the authorities.

Evidence of economies of scope is important because they indicate that it would be profitable, all else equal, to jointly produce different bank products and/or diversify into non-bank financial activities, depending on the nature of the scope economies found. US studies using 1980s data found little evidence of them, but as Table 9.2 illustrates, the results for France, Italy, Germany and Spain vary considerably. Research based on 1990s data finds more evidence of scope economies.

Pricing practices were found to deviate from textbook competition in varying respects. Early studies tended to find support for the structure – conduct – performance hypothesis, but Berger (1997) casts doubt on the results of these studies because of problems with the techniques used. The results from Berger’s study, and others, suggest the relative efficiency hypothesis may have some merit. Such a mixed bag of evidence is not much help for policy makers because the regulatory implications are so different. Until the issue of correlation between market share and concentration is properly addressed, the results of these empirical studies should be treated with caution.

Though the issue of perfectly contestable markets received some empirical attention in the 1980s, recent work suggests researchers are backing away from testing for it, perhaps because of the flaw in using the Panzer – Rosse statistic for this purpose. Intuition suggests the banking market is unlikely to be perfectly contestable, and studies finding evidence of Cournot-type behaviour (i.e. the number of firms affect pricing) must reject it. Tests using data from the UK and Canada in a generalised pricing model found considerable deviation in retail deposit and loan rates from the competitive rate, and bargains and rip-offs co-exist in retail banking markets. Policies directed at reducing consumer inertia would encourage greater competition among banks.

Studies of mergers and acquisitions address a myriad different questions such as the effects on share prices, managerial compensation, efficiency, competition and systemic risks. Takeovers do not appear to serve the interests of the bidder’s shareholders. Banks’ customers do not appear to fare particularly well either, but they may be undertaken if managers stand to gain from them. Though there are reasons for arguing that M&As could increase systemic risk, the empirical evidence is mixed. The case study approach taken by Davis (2000) tends to confirm the econometric findings, though over the longer term, i.e. 10 years, the overall effect may be positive. Even though studies of bank mergers and acquisitions may raise more questions than they answer, two important points stand out. One is the fact that merger proposals have often been disallowed when the authorities saw a serious threat to the banks’ customer interests. The other is that the fear of takeover will have constrained management and helped to deliver a higher stream of earnings than would have been expected had takeovers been prohibited. Takeovers that did not occur may have had more influence than those that did!

[ 543 ]

C A S E S T U D I E S

or ‘‘accept’’ merchants’ bills of exchange, which would be traded in the market. Merchant banks were a also called ‘‘accepting houses’’; this term was still being used to describe this type of UK bank until the early 1980s.

Though most observers have the impression of GS as being a successful firm with a reputation for steady profits, this is by no means the case. Individuals were responsible for Goldman Sachs’ history being somewhat like a rollercoaster ride – some famous names helped the firm soar to unbelievable heights but others were responsible for rapid descents in the bank’s fortunes. Though the firm established a first rate reputation in its early years, it was plagued by a series of fiascos, some of which threatened the financial viability of the firm. However, GS always managed to restore its reputation and fortunes, thanks to some key individual. This case will review the early fiascos and the individuals who rescued the firm, before turning to the mid-1980s, and the long road to becoming a listed company.

10.2.2. The Peaks and Troughs

Throughout the 20th century, Goldman Sachs suffered a number of difficult episodes, costing its profits, capital and reputation. There were numerous major incidents, followed by internal rescues.

At the outbreak of World War I, Henry Goldman supported the German cause. His father, Marcus, had emigrated from Germany in 1848. He continued to have dealings with German banks, even after the United States had entered the war. His brother-in-law, Sam Sachs, supported the English side. Nonetheless, GS lost all of its London business until after the war because Kleinworts was forbidden by the UK government to have any dealings with Goldmans.

In 1917, Henry Goldman resigned, the last of the Goldman family to serve at the bank. Though he was deemed a liability, his departure deprived the firm of a significant amount of capital and investment banking skills.4

A new partner, Waddill Catchings, joined GS in 1918, with a remit to revive the underwriting business. He helped restore the firm’s fortunes by the 1920s. However, his real interest was in trading – a number of trading accounts were opened, and GS profited from trading in foreign exchange, though they were still minor players compared to the giants such as JP Morgan.

In 1928, Catchings persuaded the other partners to enter fund management. The firm formed an investment trust, known as the GS Trading Corporation, part of a boom in investment trusts at the time. These speculative times meant the shares in the trust were soon priced at twice the value of the underlying assets.5 A second trust was created. Trusts could produce revenue from three sources: fees from underwriting and broking, the rising value of shares GS had on its own books, and new investment banking business from some of the firms with shares in the trust. Shares in GSTC were priced at $100 at issue, and peaked at $326.

4 Henry Goldman retired a wealthy man, and continued to support the German cause, planning to make it his permanent home in the early 1930s. However, the rise of the Nazis forced his return to the USA. He died in 1936. 5 Investment trusts are closed end mutual funds, structured as limited public companies but with fixed shares of capital. Ironically, one of the modern day issues is an apparent anomaly: the shares trade at a discount to the underlying asset value.

[ 545 ]

C A S E S T U D I E S

very large blocks of shares are traded on the stock exchange. Companies had started to set up pension funds after World War II, and the institutional investor was born. The lessons of the 1929 stock market crash meant these funds restricted their investments to bonds. However, the interest in investing in shares grew as the benefits of diversification became apparent.

The New York Stock Exchange was organised to serve the small investor – there was no mechanism for the huge trades institutional investors would need. Stockbrokers lacked the capital to fund larger trades. Levy agreed to buy large blocks of stock (e.g. 250 000 shares) assuming the market/price risk but confident, given Goldman’s position in the market, that it could find buyers for these shares. GS would profit from the difference between the bid – offer price and from the commissions charged for block trading.

In the late 1980s, Goldman Sachs would export its knowledge of block trading to Europe, via its London office. Up to that time, any firm wanting to sell a large block of shares had to do so through its bank, which normally agreed to act on the firm’s behalf by purchasing the block of shares at a discount of 10%, even though buyers had already been found. GS used their experience and capital to offer customers a better deal and quickly captured a large share of this market.

Levy the trader succeeded Weinberg the investment banker as senior partner in 1969. However, Levy had to report to a management committee, comprised of more senior partners, created to monitor the trader’s activities. This move illustrated the well-known animosity between traders (whose notion of the long run was a full rather than a half day) and investment bankers, who cultivated long-term customer relationships. To quote Sydney Weinberg: ‘‘I never traded. I am an investment banker. I don’t shoot craps. If I had been a speculator and taken advantage of what I knew I could have 5 times as much [money] as I have today.’’ (Endlich, 1999, p. 64)

The establishment of a management committee was not enough to stop GS from becoming involved in another major crisis when Penn Central Railroad went bankrupt in the 1970s. Penn Central’s earnings had been very poor, yet Goldman Sachs, which arranged the issue of the firm’s commercial paper, continued to sell the paper and even assured a rating agency that the company was sound. GS was confident the Federal Reserve Bank would provide the liquidity to the railroad company, and had calculated that the firm’s assets exceeded its debts. At the same time, GS was reducing its own exposure in this paper. Even though Levy claimed GS had always been confident of Penn Central’s securities, it was censured by the Securities and Exchange Commission for its activities and required to provide investors with more information in future issues. Again, GS faced a mountain of law suits, which threatened to wipe out the partners’ combined capital. Most were settled out of court but one went to a jury and GS was found guilty of defrauding its customers by selling commercial paper in Penn Central in 1969 and 1970 when the railway was nearly insolvent. The firm was forced to buy back all of the plaintiffs’ commercial paper at its initial value plus interest. However, GS also bought back paper from other clients at huge discounts. The losses were far less than expected once the value of the paper began to rise, which helped to cushion the overall losses.

Levy, with the other existing partners, agreed to inject some of their own capital to protect the junior partners from suffering capital deficits. This inspired a great deal of loyalty among members, especially as the firm grew into a much larger partnership.

[ 546 ]

M O D E R N B A N K I N G

The 1970s was a period of doldrums for the firm, largely because of the steady decline in the stock markets, which did not recover until 1980. In 1976 Levy died suddenly after a stroke. He was succeeded by two co-leaders, Mr John Whitehead and Mr John Weinberg, the son of Sidney, a partner since 1956.

Levy’s development of the trading floor left the firm well positioned to take advantage of the growth in the stock market from 1980. By 1984 there were 75 partners, each earning $5 million per year. The firm appeared to have learned from experience over time, and by this time had a well ingrained corporate culture, with the two co-heads, an eight-partner management committee and the partners. There was a clear hierarchy: senior partners, partners, vice-presidents and associates, who put in long hours in the hope of being made partner. Though there was increasing pressure for the partnership to agree to a public listing, three more scandals would delay the change until the late 1990s.

10.2.4. From Partnership to Public Listing?

In 1986, Mr John Weinberg, as senior partner, was responsible for informing those who were to be promoted from being vice-presidents to partners. The firm had grown so large that for the first time, partners did not know every candidate up for partnership. Also, many of the new partners came from trading, illustrating its importance in this traditional investment bank. These were signals that Goldmans had grown so large, it was time to look at alternative methods of corporate governance, to replace the partnership arrangement, which had served the firm since its inception. Before considering the initial attempts to get the partners to agree to a public listing, it is worth reviewing where the bank was in 1986, the year when, for the first time, serious consideration was given to the bank going public.

In 1986, the firm had four divisions:

1.Traditional investment banking: underwriting and M&As.

2.Fixed income (bond trading).

3.Equities: share trading.

4.Currency and commodities trading.7 Proprietary trading was not yet in a separate division but became one in the early 1990s.

These functions are explained in detail in the Appendix.

Until the mid-1980s, Goldmans had a well-defined, strong corporate culture. Its main features were as follows:

žA traditional recruitment procedure, with in-house training. From the early years, Goldman Sachs tended to hire from a large pool of MBAs. 1500 would apply for 30 places. The chosen few would be intelligent, mature, confident and value teamwork. The motivation of the associates and vice-presidents, in the scramble to become partner, was summarised by a senior partner, Steve Friedman, ‘‘No one ever washes a rental car’’.8 Most graduates who were hired would expect to spend their whole career with a department, so

7 GS purchased a commodities partnership, J. Aron, in 1981. 8 Endlich (1999), p. 19.

[ 547 ]

C A S E S T U D I E S

every department had a wealth of expertise. Employees were expected to give a lifetime of service. In 1984, almost all of the 75 partners had been with the firm for 10 – 20 years.

žThe firm was insular. It was rare to bring in ‘‘outsiders’’ at senior levels.

žTop priority was given to the customer, and GS was selective in the customers it chose. GS prided itself on moral standards. For example, it would not underwrite shares involving non-voting stock because all shareholders should have voting rights. Nor would they participate in hostile takeovers. As a result, any firm threatened by a possible or actual takeover sought the advice of GS. Giving priority to clients was not just to please clients. A close relationship yielded information, which in turn helped in the development of new products and services, which could be sold at a premium.

žThe management committee was the main source of decision making, though departments were encouraged to be innovative.

However, during the 1980s, there was a radical change in the investment banking business, due mainly to the growing importance of the trading side of the business, both proprietary and for clients. As a consequence, the traditional culture at GS was challenged by:

žAn increasing tendency to recruit staff at senior levels. Prominent employees from other firms had to be hired if Goldmans was to keep pace with the new parts of investment banking such as the derivative markets. By the end of the 1980s, nearly half of the GS employees had been there less than three years. The newcomers who joined as partners had little time to acquire the firm’s ‘‘lifetime’’ values. Employees of long-term standing resented the new entry.

ž‘‘Black Monday’’ (17 October 1987) forced the firm to take the unprecedented step of making a large number of redundancies. Lifetime employment was no longer guaranteed, and the loyalty inspired by it was undermined. The combination of senior level hiring and redundancies led to staff defections and early retirements. The comparatively low compensation levels also encouraged the best staff to leave for firms where compensation was linked to individual performance and short-term profitability.

žThe team approach was being undermined by all of the changes noted above. It was no longer possible to place an employee in a department for life.

žIgnoring the competition was no longer an option. For example, in 1989, Weinberg agreed to allow GS to represent a hostile bidder in a takeover attempt. Also, the firm had hoped that by taking on Maxwell as a client, it could position itself in the London markets.

Goldmans tried to address these issues by hiring consultants for advice on management techniques. Management and leadership programmes were introduced for the higher level entrants to encourage them to adopt GS’s core values of specialisation and teamwork. Though recruitment at senior level caused resentment, it also meant the firm could quickly move into new areas of the rapidly growing investment banking business.

Overall, however, there were severe pressures on the partnership system of ownership. The unlimited personal liability of partners frightened them, as the number of law suits against Goldmans and other investment banks grew. A better method of compensation was needed if Goldmans was to attract and keep the best. It could no longer rely on individuals working for years as associates and vice-presidents, especially when it began to recruit at senior level.