Figuring Asset Liability in the Global Market
During the past decade, banking has become increasingly multinational. Barriers between United States financial institutions and foreign entities were lowered or fell altogether during the Reagan years.
And banks leapt at the opportunity, opening overseas divisions to service their increasingly international commercial clientele.
Additionally, major U.S. banks, which formerly invested almost exclusively in U.S. securities, increased their investments in Europe and the Far East in search of stronger returns.
The consolidation in the banking industry in the United States has seen a great number of community banks merged into major financial institutions. And many of these big institutions have overseas operations, investments, and exposure.
How do you factor this new exposure into your asset-liability management efforts in an industry largely dependent on margins for income?
A simulation model is an important part of the asset-liability management process. However, buying a model does not solve problems of asset-liability management, interest rate risk, or currency risk; asset-liability management is too complex for a simple solution.
This is most clearly true when one looks at the interplay between interest rates and currency fluctuations. Asset-liability management is a process that can change drastically from month to month.
Don't let the input and output from your model become the definition. Managements need to understand how a particular asset-liability model relates to their institutions.
Many Factors Involved
This is difficult because the process has many components.
Asset-liability management involves concepts like simulation, transaction analysis, capital planning, regulatory compliance, multicurrency transactions, loans, borrowing, investments, rate and currency swaps, general ledger, product profitability, and market value.
Frequently all of the components come from incompatible subsystems managed by different units in an institution.
To be properly understood, these different aspects need to be divided into manageable parts.
This is considerably different from the "let's put it all in the model and see what it tells us" approach.
To understand this scattered and flawed approach, which is used by many institutions, one needs to examine the evolution of the asset-liability management process.
From Simple to Sophisticated
Asset-liability management process emerged over a considerable period of time, with different disciplines sprouting within various organizational areas.
Initially, it began with budgeting questions like: "What is most likely to happen in the next year?"
Then a planning aspect appeared. In addition to the most likely scenario, one or two other cases were considered. Institutions began to view and to protect themselves from alternative interest rate and growth scenarios.
Later, institutional planners began to shift from a macro approach to the micro. Rather than looking at the entire balance sheet, they looked at subsets or individual portfolios, asking the same questions about alternative growth and interest rate scenarios.
In the field of simulation models, we began to see more interest in larger and more charts of accounts and scenario comparisons.
Interplay of Risks and Rates
Finally, the more sophisticated institutions began to analyze risks. These institutions studied which items are affected by what kind of changes in rates.
Now, as we move toward world banking, institutions have added the capability to include transactions in different currencies to their modeling process.
The asset-liability management process in financial institutions today should cover all these aspects.
Due to the complexity of this discipline, however, the level of sophistication varies wildly from institution to institution. Often, a little bit of everything is being done without a coherent overall vision.
Merely assigning different currency conversion factors to items in a simulation model is not sufficient.
A well-run asset liability process should be able to easily distinguish the sources and magnitudes of interest rate risk from those risks that occur when using multiple currencies.
Currencies and Economies
Most economists would argue that trying to make this distinction is futile because fluctuations in both interest rates and currency conversion rates mirror the overall behavior of the individual economies.
In the ideal situation both interest rates and currency rates would move together; there would be little need to separate these risks.
This is fundamentally true -- in the long run. In practice, however, it would be difficult to state the precise relationship between a movement in the Japanese and U.S. Treasury rate spread and the movement of the yen versus the dollar.
In the real world, it takes a considerable time for the financial markets to arrive at a consensus for the time value of one currency versus another.
During this interval there is ample opportunity for both gains and losses due to mismatches in interest rates and currency.
A major difficulty in international asset liability management is the fact that common business practices differ from country to country.
In one country, time deposits may be very sensitive to interest rate movements, while in another country it's common practice to have such strong penalties that early withdrawal is practically nonexistent. Differences in taxation and tax benefits also affect customer behavior.
For asset-liability management purposes, multicurrency cost accounting and allocation become mandatory.
Money could be lent at the same currency-adjusted rate in two different countries. But due to these countries' specific factors, in addition to possible differences in purchasing power parity, there can be a tremendous disparity in the cost of making and servicing the loan.
It becomes almost impossible to correctly identify potential sources of risk when operating costs are simply totaled for all currencies.
Wholesale and Retail Banks
For a wholesale or money-center bank, it is probably sufficient to model all the institution's currency and interest rate risks in a single model, since the risks consist of multiple currency and rate arbitrages.
The retail bank, however, if it wishes to properly identify its risks, must have a separate model for each country in which it has a retail network, identify individual risks, and then have a way to combine all the data and finally consider the overall risk position.
This usually means identifying interest rate risk country by country and then factoring in the effects of currency conversions, currency swaps, and futures.