Japan’s government is finally forcing the country’s financial sector to restructure after years of ever higher nonperforming loans and terrible performance for shareholders. In 1999, it made direct capital infusions of $92 billion and allowed several banks to fail. It also indirectly suggested that Japan had room for only three or four global banks, not the two dozen it then had. In August 1999, Dai-Ichi Kangyo Bank, Fuji Bank, and the Industrial Bank of Japan (IBJ) announced plans to merge, an announcement followed by the merger of Sumitomo Bank and Sakura Bank. The number of banks had shrunk from 21 to 10 and is now down to 8.

Three means to an end

Is consolidation the answer to the profitability problems of the Japanese banking system? Mergers garner media attention and signal dramatic change. They also tend to create a domino effect, inducing other banks to respond with similar announcements. Amid this great show of activity, it is important to remember that mergers are just a means to an end—higher profits—and must be judged on that basis. In reality, a level-headed analysis of the costs and benefits of recent and prospective mergers raises many doubts. Let us look at three of the most commonly cited routes to increased profitability and see how far mergers can take Japan’s banks along them.

Economies of scale

Unit costs decline as scale increases, so economies of scale are often regarded as one of the main benefits of a merger. When you examine the scale and expense ratios of Japanese banks, you find that expense ratios go down after assets reach the level of about $28 billion but cease to do so when they exceed $92 billion. All of the remaining Japanese banks are at least this large. Moreover, few if any economies of scale result from the integration of information technology systems following a merger; on the contrary, the task is so complicated that it invariably entails further investment.

Cost cutting

In theory, the consolidation of bank branches is an effective way to cut costs. Yet the reality is often otherwise. Terminating a lease on a branch location and finding a nonbank buyer for it is difficult, for example, because the space will have been outfitted with hard-to-adapt features such as a vault. Selling a lease to another bank will become more difficult as the number of banks and branches declines.

Even if a bank succeeded in assigning its lease, the customers of the closed branch could very well take their accounts to a nearby competitor. In banking, as in other businesses, some 80 percent of the profits come from about 20 percent of the customers. That 20 percent could include a relatively high proportion of the customers of a branch that was being closed. In any case, it is doubtful that Japanese banks have any systematic way of identifying their most lucrative customers.

Mergers also create opportunities to streamline the workforce. But a glance at bank employment in the United States reveals that bank employment in Japan isn’t excessive. For example, the Bank of Tokyo-Mitsubishi has 17,937 employees; Dai-Ichi Kangyo Bank, 16,402; and Sakura, 16,381. The entity resulting from the merger of Dai-Ichi, Fuji, and IBJ will have 35,204 employees. By contrast, among US banks, Chase employs 72,683 people; Citicorp/Travelers, 173,700; and Bank of America, 170,975. That is several times more than the employment base of their Japanese counterparts, despite asset bases of comparable size.

If a Japanese bank wanted to realize significant cost savings, it would have to look not at the number of people it employs but rather at how much they earn. The average salary paid by the Japanese banks is 1.53 times the average at three US banks: Chase Manhattan, Bank of America, and Bank One. The difference is partly explained by the Japanese banks’ high ratio of executives to clerks. If pension and other benefits are considered, the ratio rises to nearly 2.00.

However, the likelihood that lower salaries would result from a merger is negligible.

Cross-selling

If the partners in a merger can cross-sell their products to current customers, profits (and those customers’ loyalty) increase as a result. US experience—the merger of Citicorp and Travelers, for example—bears this out. Before creating Citigroup, Citicorp had already made a considerable investment in managing a database of its customers, and Travelers had cross-sold property and casualty insurance, life insurance, and brokerage services. But, as the result of many years of regulation, all Japanese banks sell the same kinds of products.

Missed opportunities

Mergers should also be examined in the light of missed opportunities, for a union with other banks is no substitute for internal reform as the solution to the sector’s unprofitability. Indeed, the perception is that the larger the bank, the less susceptible it is to competitive pressure and to the possibility that it might be allowed to fail. Whether this perception is correct or not, historically the Japanese government has protected big companies. As a result, banks tend to put off urgent tasks such as the establishment of evaluation systems that reward job performance rather than loyalty and connections, the reform of bank governance, and the liquidation of stock holdings in networks of related industrial companies.

Instead of looking to mergers for comfort, Japanese banks should put their houses in order by cutting costs and making profits their top priority. A merger may sometimes be the best choice. But more often than not, it will cause management costs to soar and siphon off energy that could be devoted to difficult but essential reforms.