Long-Term Finance

Many countries are trying to formulate a strategy for reform of their financial sectors. The question that is usually posed is: "What sequence of steps needs to be taken to reform financial markets?" Instead, I think it is better to approach this issue in terms of an overall strategy. As Vaclav Havel suggested, if one is teaching someone to play chess, it is important to try to explain how to move the pieces, and to explain the strategy that is needed to win the game, rather than attempt to anticipate every essential circumstance.

Financial Reform Strategy

There is no unique way to reform financial systems. We have learned many lessons from countries that have taken similar and different routes to reform. If I were to construct a simple strategy, which could be applied to many different countries, it would begin by focusing on the banking sector. The empirical regularity is that in most countries’ financial systems the banks hold the majority of the assets. In other words, my strategy would start by addressing problems in the banking sectors.

Second, early in the reform process attention must be paid to the changes governments undertake to enable parts of the financial system to grow. Therefore, I would put emphasis on the basic infrastructure needs of the financial system, such as accounting and auditing, and on a functioning legal system, as better infrastructure helps all parts of the financial system to grow. Since, in addition to information systems, finance is primarily about contracts for the future, the ability to enforce contracts is essential to a good financial system. In fact, countries that improve the functioning of both their banking and non-banking sectors — such as equity markets — show a huge jump in their growth rates. But even countries that have developed only part of their financial sector grow much faster than those that do not develop any part of it.

Third, focus must be placed on the incentives that owners and managers of financial intermediaries face. If they are given incentives to behave prudently, that is what they will do. If they are given incentives to set up pyramid schemes, that is what will happen. For example, as mentioned in a different session of this conference, the banks in Mexico were privatized and sold for a very high multiple of their book value, and it was said that because the new owners had paid so much, they were encouraged to take risks. However, there is another angle to this: Since the owners borrowed most of the capital with only a small down payment, it did not matter if they paid half or fifty times the book value. In the end, what mattered was not the quantity, but the quality of capital, and this is what created the particular incentives that encouraged the Mexican banks to act from a short-term perspective.

Fourth, focus on developing short-term markets in the beginning of the reform process. If there are no short-term markets, it will be hard to develop long-term markets. I have yet to visit a country where there are good long-term markets and no short-term ones. And as part of the development of long-term markets, there must be significant secondary market activity. This means that traders must be comfortable taking a lot of risk. Trading long-term instruments — such as government bonds–can be very risky because they are more sensitive to changes in interest rates than short-term bonds — such as treasury bills. Traders must learn, therefore, to manage the risks associated with short-term instruments before they feel comfortable with long-term instruments.

In summary, the strategy is to begin with short-term markets, which includes developing banks and money markets. Then, to move on to intermediate instruments — such as equity — which can be viewed as either a short-or long-term instrument. (Traders issuing equities look at them as long-term instruments; holders consider them short-term instruments; equities are also easier to trade if good secondary markets exist.) In later stages of reform, long-term bond markets finally begin to develop. This later stage has not yet been reached, for instance, in many of the East Asian countries, which have only insignificant growth in their long-term instruments market.

What does it take to develop long-term lending capabilities? Where does long-term finance fit in? For decades, contributors to the development literature as well as practitioners assumed that long-term financing was not in adequate supply in developing countries, that this shortage mattered, and that it was therefore important to stimulate the supply of long-term credit. Only in the last couple of years have there been attempts to verify such assumptions. The findings of some studies in the United States, which include a lot of data on individual firms, support the hypothesis that short-term markets pay a more important role than long-term markets. These studies found that many of the firms they looked at — mostly in the manufacturing sector — were primarily financed with short-term debt, which was rolled over and extended by their bankers. Bankers liked these arrangements as it kept borrowers on a very short leash. Some borrowers even preferred them because, if yield curves were upward sloping, they would get lower interest rates with short-term debt.

Healthy firms also viewed this as a way to identify themselves in that they were willing to stand up to the risk that they might not get financing: For a manager who is really convinced of future growth opportunities, one way to identify himself is to take on more short-term debt. In a number of studies, it was found that firms which are funded mostly with short-term debt adjust faster; whenever some type of short-term shock hit them, they would adjust much faster than those companies that had been funded through long-term debt. These companies with mostly long-term debt had less pressure to rapidly adjust. However, it turns out that rapid adjustment was much better for the firms. Those firms that adjusted rapidly were the ones that survived and ended up with a higher profitability rate.

Some of these results were overturned in a recent World Bank study, which consisted of two data sets. The first set included thirty developing countries where firms were listed on stock exchanges; in many of these countries, only a few hundred firms were listed. The second set included some developing countries — such as Colombia, Ecuador and India — and three industrialized countries where 2000 or 3000 firms were listed on stock exchanges. The results of this study showed that there was a shortage of long-term finance in developing countries, after controlling for differences in the macroeconomic factors, in the adequacy of legal systems, and in the characteristics of different firms. For example, if a large firm was taken out of the developing country and put into the US capital market — everything else being equal — that firm would get more long-term credit. This was especially true for small firms which, in many developing countries, have severe limits.

In addition, our study found that this shortage really did matter. Having more long-term credit led to a jump in the productivity of firms, except when it was subsidized. In this latter case, long-term credit went to the wrong firms — that is, the ones with lower productivity and profitability levels.

These results suggests that governments should intervene to increase the supply of long-term credit, but that it must be done carefully. First, any attempt to create a long-term market will not work in a volatile macro-environment; therefore, governments must address macro-instability. Second, they must build a sound legal and other systems — along the lines mentioned above — which minimizes distortions and thus contributes to the growth of long-term financing. Third, the role of a better information system cannot be overemphasized. In this connection, the question that I would like to pose is, "Which country has the best developed ‘Islamic’ banking industry?" (Islamic banking is defined here as a financial intermediary that neither charges nor pays fixed interest, but rather pays a variable return to deposits based on the performance of its investment.) The answer is: the United States, though the financial institutions there are not called "Islamic" banks; they are called mutual funds.

These funds, which account for a large part of the US financial markets, take an equity stake in equity of their "depositors." Why then does the US have mutual funds when they are not used in countries where there is a demand for "Islamic" banking? One factor is the lack and accuracy of information in many countries which want to develop Islamic banks. In addition, most mutual funds are not engaged in any fixed interest transactions. By improving the quality and quantity of information, developing accounting and auditing systems, and improving disclosures, long-term markets and "Islamic" banks would naturally develop.

Another approach which increases the supply of long-term credit with minimal distortions is the promotion of increased use of pension funds. A great deal of attention has been devoted to this issue. In Chile, for example, there was a dramatic increase in the rate of savings after moving to fully-funded pension schemes. Although many economists agree that pension funds do not affect long-term savings rates, the evidence suggests that moving from an unfunded pension system to a funded one forces people to save and that, as a result, the savings rate increases in the short term. Once pension systems are fully funded, there should be no difference between the resulting long-term savings rate and the one that would have otherwise prevailed. But there is evidence that when fully funded pension schemes are in place, these are in effect intermediaries with very long-term liabilities which will tend to take on long-term assets to match their positions. These pension funds will foster the development of long-term markets very rapidly.

Finally, subsidies must be kept small. In directed-credit programs we studied, the schemes that worked better were those in which subsidies were small or nonexistent. The Japanese case is the best known example: After World War II, the Japan Development Bank (JDB) was established to encourage long-term industrial development. The Japanese economy had one advantage: it was large and it could stimulate any one industry through more than just one borrowing firm. The average JDB borrower received an eight-year loan with a subsidized interest rate of about 200-300 basis points over the cost of the bank’s funds. (An unsubsidized loan might have cost much more.) After the loan period, borrowers were forced to graduate to borrowing funds from commercial banks or the market. So there was a subsidy, but it was small relative to some other countries’ directed credit programs, where the interest rate was considerably negative in real terms and the subsidy could be large.

In contrast, when governments in some developing countries decide to create a steel or automobile industry because the market is so small, there is only going to be one firm in the industry. In effect, the government is telling the banks to lend to one borrower. Consequently, the banks do not invest routinely in collecting borrower information or in monitoring this credit, as they view government as the ultimate risk holder.

Lastly, going back to the Japanese example, while many think that the government intervenes heavily in the financial sector, the reality is that its policy loans account for only about 10 percent of total credit. This compares with other East Asian countries, where the share is 15 to 20 percent of total credit. Moreover, in some East Asian countries, the subsidized schemes were kept so broad-based that they were not a constraint on the banks. For instance, the Malaysia program had preferred loans for all native Malays, which did not significantly limit the banks. Indeed banks were consistently overshooting this target, meaning that it did not constitute a real constraint.

In summary, the evidence suggests that long-term finance matters, but that it is important to work with short-term markets first before moving too fast to develop long-term markets. It is also important is to encourage long-term finance indirectly. To the extent that it is done directly, attention must be paid to certain principles — namely, keeping the schemes and the subsidies limited, which ensures that borrowers graduate from them.

---