True diversification means leaving home

More than a year ago, The RMA Journal received an unsolicited manuscript on cross-border portfolio diversification and the Journal's editorial advisory board advised against publishing. A similar article has now been submitted, and a member of the board agreed to respond in a point/counterpoint.

Bank loan portfolios are built on two pillars: good underwriting and diversification. Bankers understand the merits of good underwriting--they are constantly repeating their mantra of character, capacity, condition (of the loan), capital, and collateral. But diversification is another matter; it's my contention that bankers do not pay nearly enough attention to diversification, which ultimately is or should be their highest priority. Since loan portfolios have very limited upside potential and considerable downside potential, diversification is the safety net that protects them from their inevitable underwriting lapses.

But what does it mean to be well diversified? For a regional bank, is it diversifying among many different companies in its region? For a national bank, is it diversifying among different companies in different industries throughout the U.S.? When we talk about measuring the correlation effects to compute economic capital, we are really just talking about measuring the overall level of diversification within the portfolio.

I believe diversification means packing as many unrelated things, loans in this case, in one's portfolio as possible. A question automatically arises: Are loans to companies in one region of the U.S., or for that matter loans to different companies spread over different industries in the U.S., unrelated? The answer is that they aren't. Their fortunes are heavily tied, of course, to the condition of the U.S. economy. It's a simple fact that when the U.S. economy is in recession, most U.S. companies struggle; when it is prospering, most do well.

True loan portfolio diversification requires foreign exposures. My dream portfolio would have loans to every country in the world, in every industry. I know that's not realistic and nor achievable, but it highlights just how far from ideal diversification bank loan portfolios actually are. "But, but...," I can hear you saying, "foreign lending is dangerous. What do I know about bankruptcy laws in Italy or collateral liquidation in China? Nothing."

My response is, "Yes, and what did Coca-Cola know about selling soda in Malaysia?" Coca-Cola was willing to learn because it was beneficial to learn. Bank reasons for lending overseas are generally tied to extending their reach. Meanwhile, they tend to shun the very things their clients, like Coca-Cola, are doing with great success.

Another irony is that many of these same bankers have 401(k)s, where, in the interest of sound diversification principles, they own a domestic bond fund, one or two domestic stock funds (one growth, one value), an international bond fund, and an international stock fund. Meanwhile, their bank's loan portfolio is almost exclusively domestic.

These are not idle, abstract considerations. They have real consequences for economic capital. Banks keep more economic capital than they need to--a lot more--because of their reluctance to diversify overseas. The software that computes credit capital sees the domestic concentration and penalizes the portfolios accordingly with extra capital.

Also, economic capital increases with the unpredictability of loan losses. Long periods with low losses and a year or two with big losses (such as in the U.S.) are characterized by more uncertainty and unpredictability and therefore require extra capital. Better to have regular losses, with no big losses and no small losses. There is less variability in this scenario and, all things being equal, less need for economic capital to protect the bank. And, as we all know, Wall Street loves earnings predictability.

How can we achieve a stream of more predictable losses; conversely, how can we avoid surprisingly high losses? By diversifying our loan portfolio with loans to countries with low correlation to our business cycle. When their economies are in recession, ours often will not be; when our economy is struggling, theirs probably will not be. Even among trade-interdependent countries, the timing of expansions and recessions is different--witness Japan and the U.S. over the past decade.

The good news: Given the huge concentration of U.S. loans in most banks' portfolios, banks are basically in a "free lunch" state when it comes to adding any foreign exposure. Because of the big diversification benefits, they can potentially expand their loan portfolios with foreign loans while shrinking economic capital. Of course, they eventually could reach the point of diminishing returns by adding too many foreign loans, but few are remotely close to that possibility yet.

A few years back, the consulting firm McKinsey & Company did an analysis and concluded that swapping 25% of a loan portfolio into similar foreign loans reduced capital by a whopping 31%. That's equivalent to boosting a bank's RAROC return by 45%. They also noted at that time, "Several North American banks have built partnerships with international banks to swap undesired risks in an effort to reduce concentration risks."

Below, for example, is a table that shows the overall potential improvement in a bank's aggregate loan RAROC based on percentage reductions in economic capital. Only foreign lending offers the ability to reduce economic capital by these kinds of dramatic numbers.

Reduce economic  Boost loan
  capital by:    RAROC by:

      10%          11.1%
      20%          25.0%
      30%          42.9%

Now, to state the obvious, all foreign countries are not created equal when it comes to risk and return, nor are they created equal when it comes to diversification. For a bank just starting out, it seems prudent to begin with just the industrialized countries, where the economies are stable and there is a history of respect for the rule of law and the priority of lenders.

The eye-popping returns, however, are to be found in the emerging world. I would liken emerging-world lending to consumer or subprime lending -- default rates are going to be high (for example, Argentina), but such risk is also associated with appreciably higher yields and growth.

There are clear diversification gradations even within the industrialized nations. Based on one bank's portfolio of a few years ago, KMV Corp.'s Portfolio [TM] attributed the least capital to Switzerland, Japan, Sweden, Germany, and France, in that order. All things being equal, loans to those countries would have required a fraction (20-50%) of the economic capital that the same loan in the U.S. required.

A bank's goal should be to maximize its return on capital. Diversification is basically a packing strategy toward that end. When I say a packing strategy, I mean packing the greatest number of loans into the portfolio for a given level of economic capital. Like rearranging the items in a suitcase to squeeze more in, diversification allows you to rearrange your portfolio to pack one or two more loans into each economic capital dollar. Diversify some more and you can pack in a few more, and so on. The beauty of this approach is that the portfolio and its profitability expand simultaneously. The bank is growing its way to greater efficiency.

Given corporate bond rates at that time, all of those countries except Japan looked more attractive on a risk-adjusted basis than the U.S. And, as you can see in Figure 1, this is independent of industry. Any industry we choose in these countries requires less capital than its U.S. counterpart.

The reality these days is that the top companies operate globally. Coca-Cola operates in virtually every country in the world. Citigroup operates in 100 countries. American International Group, one of the most highly regarded financial services companies in the world, derives 50% of its revenues outside the U.S. American Express operates worldwide. GE Capital, which accounts for 50% of behemoth GE's revenue, operates globally.

They should also bear in mind that investors find it very easy to diversify their holdings domestically. Most have heeded the call to diversify by buying companies from different industries. Diversifying geographically, however, is more difficult for them. They can either buy international mutual funds or invest in local companies that are already internationally diversified. This explains why internationally diversified companies generally are rewarded with above-average P/E ratios. In a recent analysis of the benefits of diversification, Rory F. Knight, dean of Templeton College, Oxford, concluded: "A company appears to experience a significantly greater price-earnings ratio, for example, under a strategy of geographical diversification than does a purely domestic company."

Increasingly, the marketplace is global for financial services companies, and U.S. banks need to keep that in mind as they go forward so as not to be outmaneuvered.

There is an amazing level of disrespect accorded U.S. banks in the stock market. They have some of the lowest P/E ratios in the marketplace. U.S. banks need to take advantage of every opportunity at their disposal to boost income and to reduce economic capital. Diversifying internationally is one such glaring opportunity--an opportunity most U.S. banks have ignored to their and their investors' detriment for far too long.

[FIGURE 1 OMITTED]

RELATED ARTICLE: Counterpoint: Good Risk Management Tells Us to "Stick to Our Knitting"

Dave Lobell correctly points out in his article that in addition to good underwriting, diversification is imperative for proper loan portfolio management and in general bankers do not pay enough attention to diversification. He then goes on to state that merely diversifying the loan portfolio by making loans outside a banks region or in different industries is not enough. A loan portfolio needs foreign exposure for true diversification.

While his points about the benefits of diversification are right on the mark, his conclusion that banks should lend overseas is flawed for several reasons:

* Making loans isn't like buying shares of stock. You cant just wire money to a foreign borrower. To loan money prudently, you must, in addition to studying the financial statement, visit the company and meet its management; also, you must regularly keep up with the company through calls and visits. To do this means having a presence in each country where you loan money, and this is feasible for only the largest multinational banks and in the most developed countries.

* When you lend to Sony or Erikson or BP Petroleum, you aren't lending in Japan, Sweden, or Great Britain because they are just as multinational as Boeing or IBM or other American multinationals. To get geographic diversification, then, you would have to lend to smaller local companies, which is difficult to do as a foreign bank even with a presence in the country.

* Recent history has shown that most developed countries move in tandem with the U.S. economy-"the U.S. sneezes and the rest of the world catches a cold." It is therefore doubtful that significant increased diversification of risk could be obtained even if it were practical for a regional or community bank to lend funds outside the U.S.

Banks will never be able to avoid all the vicissitudes of the economy in their loan portfolios, but through good diversification of products, industry, geography, and size of transactions, it is more likely that no one event will prove catastrophic. This can all be accomplished without venturing outside the U.S., where most banks know little or nothing about the risks or the economies.

Having said this, I do not wish to diminish the validity of Lobell's argument that diversification is as important or more important than good underwriting. Diversification, in my view, should be accomplished by banks "sticking to their knitting"--that is, lending money in reasonable amounts compared to their capital to companies and individuals they know. Diversification needs can be met by selling participations of larger credits to banks who share their philosophy and underwriting principles.

Raymond Rusnak

Senior Credit Officer

LaSalle Bank N.A.

ray.rusnak@abnamro.com

Lobell is a member of the Treasury group of a large Illinois bank, where he is working on Basel II capital preparations. He was formerly with First Union, now known as Wachovia Corporation, where he worked with various portfolio models. Lobell's views do not necessarily reflect those of his employers.

COPYRIGHT 2002 The Risk Management Association
COPYRIGHT 2005 Gale Group