Expanding Networks Globally Is a Balancing Act

Second of Two Parts

Deploying information technology beyond national borders-a task that is becoming a requirement for an increasing number of institutions-presents a series of challenges.

The world beyond the Americas, Europe, and Japan presents its own unique characteristics. China, Eastern Europe, and the countries of the ex-Soviet bloc were denied advanced technology for so long that they have engaged in leapfrogging since about 1990. They have had to build whole sets of infrastructures all at once, including basic accounting, account processing, and payment systems.

Much of this technology has come from Western systems integrators or hardware vendors. The World Bank has funded many efforts, recognizing that modern economies cannot exist without banking systems. The predilection has been to get hardware, software, and services from as few sources as possible. Some disasters have resulted. Czech banks, for example, had never used checks, double-entry bookkeeping, standing orders, or direct debit. Several lawsuits are still pending-the most famous of which is Ceska Sporitelna v. Unisys, for more than $100 million.

The low-cost, efficient telecommunications capacity that we take for granted in the United States is almost wholly lacking in the underdeveloped world. State Bank of India, for example, has $50 billion of assets and 230,000 employees. But only 3% of its of 4,000 branches are networked. In other countries, electronic connectivity is erratic or limited in bandwidth, and often makes centralized, high-volume processing unreliable.

Of course, product features and functions vary widely. Islam, for example, prohibits charging interest. Banks offering "Islamic finance" must use discounts or fees instead. In Central Europe the use of revolving credit cards is limited because overdraft facilities equal to three months' salary are automatically extended to current account holders. Unlike any other form of consumer credit, repayment of these facilities is covered by government-sponsored social insurance programs.

The effect of these national differences on the application systems can be extensive. High flexibility is required to change code, data, and processes. At the same time, building blocks such as debits, credits, and interest calculations have a high degree of universality. The trick is finding a business approach that meets local variations and trends without reinventing the wheel in each country.

In China, private homeownership was allowed only recently; much housing is provided by an employer or university for as little as 4% of monthly salary. So banks are pioneering a mortgage market, which totals about $1.5 billion today. Chinese banks can be important new customers for loan origination and servicing applications.

Elsewhere in Asia, deregulation in various forms will allow new competitors, new products, and an acceleration in new processes and channels. IT spending will probably be stimulated once the current constraints are overcome.

Asian markets still funnel most capital flows through bank balance sheets, obscuring credit conditions and increasing systemic risk. The solution is to stimulate faster growth of capital markets, which will in turn trigger expanded growth of market data, trading technology, risk management, exchanges, and settlement and clearing systems.

This diversity will not change. Consolidation will occur, and this will alter the solutions and resources available. But as fast as some products, services, or technologies converge, other issues, needs, and products will emerge.

For vendors, having a global presence will become mandatory. Traditional reliance on proprietary applications will decrease. U.S. vendors must adapt their solutions to a multitude of local needs and therefore make products extremely flexible. For non-U.S. vendors, true global success requires penetrating the U.S. banking market, with all of its unique attributes and intense vendor competition.

Hundreds of successful vendors already operate globally. The closer the technology is to capital markets or to enterprise challenges, the more common this is. Algorithmics, the Toronto-based risk management vendor, obtains only 2% of its revenue from Canada. The closer the technology is to retail banking, the less likely is global coverage.

Wholesale banking technology is somewhat in the middle, as that business does have common characteristics around the world. For example, large global corporations want to be able to pay all their partners through one bank, yet most countries have closed proprietary systems that provide cash management services. Integrating these proprietary systems is difficult, and even global banks can integrate only a few of their largest clients.

Financial institutions have three potential strategies. Many, but not all, have grown internationally by acquisition. The most common strategy therefore is "leave it alone." Separate subsidiaries or banks in different countries would simply provide their own technology. For example, most European and Japanese banks that have bought U.S. banks have made few changes in their IT.

The advantage of this is low cost. However, what is the point of operating globally without common products, data, services, or channels? Some of this can occur even with completely separate systems, but costs remain high, and global initiatives will occur at a snail's pace.

We might call the second strategy "global best of breed." American Express and Citicorp, for example, have brought together chief information officers from various subsidiaries to share ideas and solutions in an internal free market. BankBoston has worked to export its expertise in retail banking technology to its Brazilian and Argentine subsidiaries.

Some pieces of technology may be determined to be suitable for use in multiple countries. Certain standards may be promulgated, such as the use of hardware platforms, languages, or protocols. Certain IT operations such as networks and data centers may be centralized or regionalized. Global relationships with key vendors can be established and monitored for better control.

The last strategy would be "enforced similarity," an effort to dictate common systems everywhere. This approach has the most potential for reduced costs, common core products, and equal treatment of customers everywhere. It also has the greatest risk.

Various manifestations of "global best of breed" are probably the best strategies today for serious players. They can begin the process of exploring a world perspective. But management must be prepared. Owning multiple large banks in several countries may result in none of the existing solutions being appropriate. New platforms may be needed, particularly when countries with low transaction volume are involved. Ambitious efforts to provide cross-border data and products may be admirable, but they are expensive.

Over time, globalization will be more unsettling for U.S. banks. The United States is such a large market that some U.S. banks can reach or exceed $50 billion of assets and have virtually no global experience. Increasingly, this isolation will become untenable. A path to the world's financial markets is needed and so is a gradual plan for finding the needed technology. Those that do the best job at this will maximize their chances for success.

For reprint and licensing requests for this article, click here.
MORE FROM AMERICAN BANKER