Limitations of a priori criteria

Before discussing the potentialities and limitations of interest rate policies, it is pertinent to refer to some conceptual problems involved in the selection of criteria to assess the effectiveness of interest rate policies as well as in the choice and interpretation of interest rate statistics as a basis for policy making.

Since capital is scarce, it follows that there should be an incentive for saving as well as a device for its efficient allocation among alternative uses. These twin functions are discharged, although with varying efficiency, through the interest rate mechanism. But even a perfectly competitive capital market does not ensure an optimum level of savings or a socially desirable pattern of investment. In the less developed countries, where capital markets are subject to even more imperfections than in developed countries, the structure and level of interest rates as determined by the free play of demand and supply do not always reflect the true economic cost of capital. This raises the question of what, if any, the rational and practical criteria are for an interest rate policy in the developing countries.

Although many attempts have been made to derive a conceptual rate of interest, variously defined as the “accounting” or “shadow” rate of interest,⁸ as a theoretical norm with which to measure the deviation of actual rates of interest, their practical utility is questionable. First, such models are essentially aggregative, static equilibrium models based on highly restrictive and unrealistic assumptions, such as a closed economy, perfect competition, production functions with constant returns to scale, constant savings ratios, constant marginal propensities to save, the absence of state interference, and the lack of technical progress. All these assumptions are scarcely relevant to the problems of developing economies, which have to accumulate capital at an increasing rate to attain self-sustaining growth. The assumption of constant savings ratios is particularly restrictive, as it abstracts from the crucial relationship between the savings function and the levels of income and interest rates and its bearing on the process of growth. Second, such models raise difficult empirical problems of measurement of the relevant variables and parameters, such as the stock of capital, output/capital ratios, factor shares, savings ratios, which are by no means easy to estimate in the less developed countries because of the paucity of basic data. Moreover, they yield a unitary rate of interest⁹ for the whole economy that, however elegant theoretically, can scarcely function as even an approximate index of the real cost and availability of savings in the highly fragmented capital markets with the multiplicity of customer interest rates that is typical of the less developed countries. It is therefore hardly surprising that even the most sophisticated development plans in Asian countries have not used shadow rates of interest. These criticisms of the operational significance of shadow rates do not, however, detract from the heuristic value of such models. They can serve as broad guidelines for more efficient allocation of resources, particularly if a system of accounting prices were adopted for valuation of all major scarce inputs, such as foreign exchange and skilled labor.

The impracticability of identifying optimum rates of interest also reflects the more fundamental problem of finding objective criteria for “optimum” saving¹⁰ or “optimum” investment, because a major issue in the choice of the optimum rate of saving is one of equity between the present and the future, and this cannot be resolved without recourse to temporal value judgments. Obviously, if it is difficult to optimize savings or investment without value judgments, the same objections would apply to optimum rates of interest.¹¹

Some possible empirical criteria

Since there are no operationally meaningful a priori criteria of what constitutes a realistic and appropriate level and structure of interest rates for any economy, policy in this respect is necessarily reduced to a matter of judicious empiricism, keeping in view the objectives that are sought to be achieved through regulation of interest rates. Therefore, interest rates are best regarded as “multivalued instruments” rather than as “targets” of economic policy. The appropriateness of interest rates therefore must be judged strictly in relation to policy objectives, depending on the aspect of interest rates that is to be assigned the greater weight, i.e., their possible role as an incentive to financial savings (income factor) or as an instrument of credit policy (cost factor). Although policy has to be based on a delicate balancing of the multiple role of interest rates, the present paper is concerned largely with their potentialities in mobilizing financial savings in the organized sector.

In this context, a useful but not wholly unambiguous criterion is to examine whether the money rates of interest on typical savings media, such as bank deposits and government securities, are positive in real terms after deflating their nominal levels for changes in the purchasing power of money as measured by some appropriate index. But the significance of the resultant real rate would depend on the indices used, the terms and maturity of the interest-bearing asset, etc. The cost of living index, being more relevant to savings decisions than are variations in the wholesale price index, would be a fairly satisfactory index. But the criterion of ensuring a positive real rate of interest, which is exceptionally significant in periods of high or rapid inflation, ought not to be applied uncritically regardless of prevailing conditions. A substantial volume of financial saving takes place irrespective of whether the real rate of return is positive or negative, because of a variety of factors, such as the money illusion, the minimum demand for cash balances, and the captive market for government securities. Moreover, a rigid adherence to the yardstick of positive real rates of interest may necessitate the manipulation of money rates every time the rate of price change altered and when the upward change was sufficiently large to bring the real rate below zero. Such frequent changes in money rates are neither desirable nor feasible and may even have unsettling effects on the propensity to save in financial media, even if the frequency of change in money rates could be diminished by using a moving average of current and recent rates of inflation. A rational interest rate policy therefore has to be predicated on the basis of fairly stable expectations for reasonable periods of time.

Insofar as expectations of rising, not steady, prices become normal, the effects of even high nominal rates of interest may be dampened. For instance, if prices are confidently expected to rise at 3 per cent per annum, a nominal rate of interest of 6 per cent would be the same in real terms as a rate of 3 per cent. Consequently, given sufficiently inflationary expectations of rising prices, even a relatively high nominal rate of 6 per cent would be regarded as merely normal. The nominal rates will therefore have to be raised sufficiently high if they are to exercise any restraining effect and to ensure consistently positive real rates of return on financial assets. A policy of ensuring positive rates of return through the manipulation of nominal rates has several advantages over “indexing” devices. It obviates difficult technical and administrative problems of linking the nominal value of savings media to some index, such as prices, gold, or foreign exchange, since it operates through the price mechanism. It does not carry any implicit official acknowledgement of the inevitability of inflation and the incapacity of the authorities to control it. The successful experience of China, Korea, and, more recently, Indonesia in ensuring positive real rates of interest through manipulating nominal rates rather than by resorting to indexing techniques testifies to the merits of this approach to a savings strategy, at least in developing economies. The indexing technique, significantly, has been found more practicable in comparatively sophisticated economies, such as Finland and Israel, although it is arguable that even in such economies an interest rate policy could be more efficacious. But, if pursued in isolation, even a positive interest rate policy might be open to the objection, equally applicable to indexing devices, that for any economy to seek to protect the value of one category of income (returns on financial assets) without similar protection to other categories (such as wages or rent) would be inequitable and inefficient. This, however, should be regarded as arguing the need for a coherent and comprehensive incomes policy rather than as a demerit of an interest rate policy.

The broad criterion of assessment of interest rate policies adopted in this paper is their efficacy in mobilizing voluntary private (financial) saving. It will be argued that the level of interest rates can help to influence rates of financial savings in developing countries and probably total real savings in the national accounts sense. Insofar as there is a critical range within which interest rates could be said to be effective in evoking larger financial savings, there is a strong case for a flexible policy aimed at a realistic level of rates high enough to stimulate saving without constricting the rate of development. By maintaining interest rates at unrealistic levels, developing countries may thereby deny themselves the use of a potent instrument of policy. On the other hand, interest rates need not always be high and rising or pegged at a given level. Rather, a minimum objective should be to ensure a positive real rate of return to the saver and, as long as this condition is satisfied, the level and spread of nominal rates can vary in keeping with other requirements of policy. In fact, the choice is not so much between once-for-all high or low interest rate strategies as between rigid and flexible policies. The monetary authorities would therefore need to raise or lower interest rates steadily and progressively over a period of time. It would therefore not be enough to establish a new level of interest rates and then merely seek to maintain it. There might, however, be occasions when a policy of stable nominal rates might be both desirable and feasible. It is, however, difficult to measure and predict the magnitude and speed of response of the volume of saving to interest rate changes, even if it were possible to draw broad inferences regarding the direction of the impact. In less developed economies there is likely to be a considerable dampening of the speed and impact of changes in capital values (interest rates) on savings behavior because of the low ratio of financial assets to gross national product (GNP).¹²

Likewise, interest rate changes are more likely to affect rates of financial saving and therefore their net impact on aggregate real savings cannot always be known and identified. Even for interest-bearing financial assets, interest rate changes may merely change the form of saving without affecting the total, as, for instance, when savers decide to “switch” from idle cash balances into time deposits or from the latter into government bonds. Since it is more important to increase total net savings, the forms that they assume are relatively subsidiary. Thus, the mere holding of idle balances (currency or demand deposits) is as much an act of saving as the holding of near-money assets and bonds, as both involve the decision not to spend and to release real resources.

On the other hand, the forms that savings assume are of consequence from other points of view. The flow of savings into interest-bearing financial assets in the organized sector facilitates a more efficient allocation of savings between competing uses by subjecting them to the discipline of the price mechanism of the money and capital market; it also enables the monetary authorities to exercise better control over the asset preferences of the financial system and of firms and households, since financial assets are more sensitive to changes in interest rates. Savings held in idle balances are liable to be dishoarded more easily than financial assets. Therefore, there are net advantages in a transfer of savings into interest-bearing financial assets, particularly for the less developed economies. From this point of view, a purposive interest rate policy has to seek to extend the organized sector relative to the unorganized sector by, among other things, maintaining interest rates within a “critical area,” i.e., one within which financial savings and probably total savings are responsive to changes in the structure and level of interest rates. In fact, a good test of policy is to examine whether the interest rate weapon has been fully used within these limits.

Among other possible criteria of the appropriateness of interest rates in the organized sector of the less developed countries are the levels in (1) the unorganized sector in the same country and (2) developed economies. But the rates in the unorganized sector are not meaningful for evaluating rates for organized finance, since the former contain substantial elements of monopoly profit, high-risk premiums, and excessive costs of administration. Moreover, as one of the prime objectives of policy in the less developed countries is to reduce the usurious levels of interest rates in the unorganized sector, it is hardly appropriate to invoke them as a basis for judging levels in the organized sector.

Comparison with rates in developed money and capital markets abroad is perhaps more meaningful if the less developed country in question is an open economy with freedom of movement of short-term and long-term funds. But even in those countries that have assumed the obligations of Article VIII of the Fund’s Articles of Agreement, such as Malaysia and Singapore, there are controls on the transferability of funds on capital account. Short-term rates on treasury bills and commercial paper abroad in competing markets—such as London, where domestic funds have freedom to move—are much more relevant. It is, however, a matter of judgment as to how far rates abroad should influence domestic rates. Moreover, it is arguable that capital movements are also influenced by speculative and confidence factors as well as interest rate differentials.

While the level of rates abroad may not have much direct relevance for domestic interest rate management for economies with almost wholly insulated national money and capital markets, such as India and Ceylon, it may well furnish a useful guideline, inasmuch as rates in capital-scarce countries, prima facie, would need to be higher than comparable rates in developed countries with higher rates of financial saving. This is, however, a complex theme, which is examined critically in the next section of this paper.

Limitations of interest rate statistics

Any discussion of interest rate policies has necessarily to take account of the limitations of existing information on the structure and level of interest rates. In the absence of adequate data on the structure and behavior of interest rates in the unorganized sector, as distinct from rates that merely help to convey some representative “orders of magnitude,” the scope of this paper is limited mainly to interest rates in the organized sector. Even in this sector the published data relate either to minimum or maximum interest rates or to some representative rate. But in order to arrive at the actual effective rates, nominal rates (for lending rates) have to be appropriately adjusted upward for (1) compensatory balance requirements, which often range as high as 20 per cent and (2) commission charges and commitment fees. These items (which vary from bank to bank and sometimes even from customer to customer), in conjunction with the nominal interest rates, have the effect of raising the actual rates to a point that is even higher than the permissible maximum. Moreover, for a more precise evaluation of the significance of quoted interest rates, one would need a weighted average of lending rates, with the weights being assigned according to the volume of lending at each quoted or customer rate of interest. Such data, however, are virtually impossible to collect, and it is consequently difficult to arrive at a “weighted” average of loan rates of interest. Moreover, “not only is there a whole range of interest rates, but also a variety of conditions which may be attached to a loan. Each transaction, therefore, has its unique characteristics. As a result, for each borrowing operation there is one particular rate of interest, which is the Relevant Rate of Interest.” ¹³ There is therefore in the real world a spectrum of interest rates with credit being rationed at each level.

On the other hand, for deposits the nominal quoted interest rates usually represent the rates actually paid to the customer, except in rare cases, such as call money deposits, where prime customers may get slightly more favorable rates than those published. But, by and large, finely quoted customer rates are more characteristic of loan rates than of deposit rates. Therefore, on the whole, published rates on deposits are a better approximation to actual rates than are loan rates.

But the information on deposit as well as bond rates has to be suitably qualified to take account of such factors as premiums, tax privileges (e.g., exemptions), and other advantages from which depositors and bondholders may benefit. For marketable government securities the extent of official support is an important factor, since such support, whatever its other merits, implies that the resultant rates are not strictly free market rates. To that extent they do not represent “pure” rates of interest. But despite this limitation, market yields on government securities represent the nearest approximation in any economy to a pure rate of interest as measured by the obligation that is basically risk free and most widely held.

Instruments and rationale of interest rate policies

An interest rate policy may be defined briefly as any official action designed to influence the level and structure of money rates of interest through statutory means, money market intervention, or moral suasion to attain given ends of credit policy and to help in the mobilization of saving through financial media. Such action may comprise (1) statutory ceilings (e.g., usury laws) and regulation of moneylending and pawn-broking; (2) statutory or voluntary interbank agreements on deposit and loan rates; (3) open market operations and bank rate changes; (4) subsidization or regulation of specific rates (e.g., on small savings, export or housing finance, rural credit). In most Asian countries one finds a combination of these instruments applied with varying degrees of efficacy. Usury laws and their ancillary apparatus of regulations are aimed chiefly at the unorganized sector and are therefore left out of account in the present discussion. The two other categories of policy instruments fall largely within the purview of central bank action designed to influence interest rates in the organized money market by changes in the bank rate or in open market operations and/or by interbank agreements. Because of the undeveloped character of the money and security markets in most of these countries and the limited reliance on central bank credit, both open market operations and bank rate changes have limited usefulness. Consequently, in practice, official regulation mostly takes the form of interbank agreements, either voluntary or statutory. This takes into account the statutory powers of central banks to regulate directly the rates on deposits and loans.

But “it is largely as a result of introducing social considerations that the policymakers face a real dilemma, when it comes to the choice of an appropriate level and structure of interest rates.” ¹⁴ In fact, the case for a pragmatic approach to interest rate policy based on modification of the market price mechanism by extramarket criteria is even stronger in the less developed countries, owing to the greater imperfections of the money and capital markets and the stronger element of officially determined priorities in the allocation of capital funds.

In the light of the foregoing considerations, we may first examine the empirical evidence on the structure and level of interest rates in Asia, particularly with a view to finding out whether there is any substance in the oft repeated observation that interest rates in the organized sector of the less developed countries tend to be low and rigid. The subsequent sections review the experience of (1) high interest rate strategies in China and Korea and (2) conventional interest rate policies in Malaysia and Singapore.

Are interest rates low in less developed countries?

The organized sector (a critique of the Myrdal thesis of low rates)

In the less developed countries the level of interest rates in the organized sector (comprising joint-stock banks, other financial institutions, security markets, etc.) tends to be substantially below that in the unorganized sector. They are almost the same as, or lower than, similar rates in the developed countries, where savings are much higher; and even where the rates are slightly higher than in the developed countries, it could be argued that the gap does not fully reflect the relative scarcity of capital that is normally characteristic of such economies. To that extent, the rates are adjudged to be low and therefore unrealistic and inappropriate. Often such judgments tend to partake of obiter dicta not substantiated by any detailed analysis or empirical investigation.²¹ A conspicuous exception is Myrdal’s analysis ²² of the inappropriateness of low interest rates in the organized sector, which, being among the most substantial of its kind, merits critical appraisal as representative of this point of view even though it was made in the context of the insulated credit and capital markets of India, Pakistan, and Ceylon. Its main propositions may be summarized as follows:

Despite the great scarcity of capital in the less developed countries in comparison with the developed countries, it is paradoxical that rates of interest, despite recent increases, are conspicuously low and in fact not as high as in many developed countries. This paradoxical policy has been sustained by a combination of selective and discretionary controls on credit and investment and by concessional finance, and is sought to be justified on the grounds of inappropriateness of high interest rates in developing countries and the insensitivity of investment to interest rate changes in a highly protected economy.²³

On balance, according to Myrdal, a substantially higher level of interest rates in conjunction with a dismantling of discretionary controls would be more in harmony with the prevailing scarcity of capital in the South Asian countries and would induce economy in the use of capital. It would stimulate greater inflow of foreign capital and mobilize more domestic saving in productive forms instead of being dissipated in speculation, hoarding (particularly of gold), conspicuous consumption and investment, and purchase of land (at inflated prices) and foreign securities. It would also bring about a decline in oligopolistic profits and in capital and land values; the latter would enhance the feasibility and reduce the cost of large-scale changes in land ownership and tenancy. Since the major portion of government debt is intragovernment debt, higher interest rates would only bring about a change in accounting relationship within the public sector. Insofar as bonds are held by private persons, higher rates would tend to bolster the rentier class but higher incomes from this source would be easy to trace and to tax, even if income from loans in other forms would, as now, be more difficult to trace.

We may now examine critically the validity of the Myrdal thesis in the context of Asian economies and its policy implications.

A comparison of the relative range of interest rates (Table 2) in the developed and less developed countries suggests that Asian countries are in two clearly defined groups, namely, Group I (Burma, Ceylon, India, Malaysia, Nepal, the Philippines, Singapore, and Thailand), which has more or less conventional rates of interest that are about the same as or lower than those in the developed countries, and Group II (China, Indonesia, and Korea), with interest rates that are far above those in the developed countries or other less developed countries. In fact, the nominal rates in the organized sector in these three countries approximate those in the unorganized sector of the countries that have low interest rates.

Table 2.

Developed and Less Developed Countries: Comparative Rates of Interest, 1958–69

(Range in per cent per annum) ¹

Sources: International Monetary Fund, International Financial Statistics; national sources.

¹Approximate range between low and high figures.

²1969.

³One year.

⁴Postal.

⁵Overdrafts.

⁶Post Office and Trustee.

⁷Seven days’ notice.

⁸Prime rate.

⁹Three months.

¹⁰Certificates.

¹¹October 1969.

¹²1949-69.

¹³Rediscounts.

¹⁴Since October 1968.

¹⁵1968.

¹⁶Private banks.

¹⁷1961–69.

Table 2.

Developed and Less Developed Countries: Comparative Rates of Interest, 1958–69

(Range in per cent per annum) ¹

Country			Call Money	Treasury Bills	Savings Deposits	Time Deposits	Long-Term Government Bond Yields	Central Bank Rate	Commercial Bank Advances
Developed countries
	France		3.51–10.41	—	4 2	4.50 2, 3	4.94–6.80	3.50–8	—
	Japan		—	—	3.60 2, 3	5.50 2, 4	—	5.48–8.40	7.40 2, 5
	United Kingdom		—	3.37–7.89	2.502, 6	5 2, 7	4.82–9.46	4–8	8½2, 8
	United States		—	1.84–5.35	4 2	5 2, 3	3.43–6.51	2.50–6	—
Less developed countries
	Group I
		Burma 2	—	—	0.50 9	1¼3	4.5 10	—	4–6
		Ceylon	—	—	3½ 2	—	2.70–5.36	2.50–5.50	5½–10
		India	1.85–6.28	—	3½–4 11	—	4.05–5.58	4–6	—
		Malaysia	3–5⅜ 2	42	3½2	6 2	6.25 2	—	7½–8
		Nepal	—	—	4½ 2	6–7 2	—	—	7½–9
		Singapore 2	—	4	3½–4	6 3	—	—	7½–8
		Thailand	6–12 2	5 2	3½2	—	7 2	5–8	9–14
	Group II
		China 12	—	—	5.40–549	9.72–243	—	10.80–14.40 13	13.32–118.80
		Indonesia 14	—	—	1.5–369	2.5–723	—	—	84–150 15
									48–120 2, 16
		Korea 17	—	—	16.8–309	15–26.43	—	10.22–23	18–28 5

Sources: International Monetary Fund, International Financial Statistics; national sources.

¹Approximate range between low and high figures.

²1969.

³One year.

⁴Postal.

⁵Overdrafts.

⁶Post Office and Trustee.

⁷Seven days’ notice.

⁸Prime rate.

⁹Three months.

¹⁰Certificates.

¹¹October 1969.

¹²1949-69.

¹³Rediscounts.

¹⁴Since October 1968.

¹⁵1968.

¹⁶Private banks.

¹⁷1961–69.

View Table

Table 2.

Developed and Less Developed Countries: Comparative Rates of Interest, 1958–69

(Range in per cent per annum) ¹

Country			Call Money	Treasury Bills	Savings Deposits	Time Deposits	Long-Term Government Bond Yields	Central Bank Rate	Commercial Bank Advances
Developed countries
	France		3.51–10.41	—	4 2	4.50 2, 3	4.94–6.80	3.50–8	—
	Japan		—	—	3.60 2, 3	5.50 2, 4	—	5.48–8.40	7.40 2, 5
	United Kingdom		—	3.37–7.89	2.502, 6	5 2, 7	4.82–9.46	4–8	8½2, 8
	United States		—	1.84–5.35	4 2	5 2, 3	3.43–6.51	2.50–6	—
Less developed countries
	Group I
		Burma 2	—	—	0.50 9	1¼3	4.5 10	—	4–6
		Ceylon	—	—	3½ 2	—	2.70–5.36	2.50–5.50	5½–10
		India	1.85–6.28	—	3½–4 11	—	4.05–5.58	4–6	—
		Malaysia	3–5⅜ 2	42	3½2	6 2	6.25 2	—	7½–8
		Nepal	—	—	4½ 2	6–7 2	—	—	7½–9
		Singapore 2	—	4	3½–4	6 3	—	—	7½–8
		Thailand	6–12 2	5 2	3½2	—	7 2	5–8	9–14
	Group II
		China 12	—	—	5.40–549	9.72–243	—	10.80–14.40 13	13.32–118.80
		Indonesia 14	—	—	1.5–369	2.5–723	—	—	84–150 15
									48–120 2, 16
		Korea 17	—	—	16.8–309	15–26.43	—	10.22–23	18–28 5

Sources: International Monetary Fund, International Financial Statistics; national sources.

¹Approximate range between low and high figures.

²1969.

³One year.

⁴Postal.

⁵Overdrafts.

⁶Post Office and Trustee.

⁷Seven days’ notice.

⁸Prime rate.

⁹Three months.

¹⁰Certificates.

¹¹October 1969.

¹²1949-69.

¹³Rediscounts.

¹⁴Since October 1968.

¹⁵1968.

¹⁶Private banks.

¹⁷1961–69.

In terms of comparative nominal interest rates, Myrdal’s generalization would therefore appear to be valid only for some countries in Group I. Even within this group it would be misleading to generalize about countries in which financial saving shows varying degrees of response to more or less similar levels of interest rates. For instance, even with comparatively low and stable nominal rates of interest the growth of financial saving in Malaysia and Singapore, as shown in a subsequent section, has been impressive partly because the real interest rate has been positive, owing to prolonged price stability and the resultant confidence of the investor in financial assets. It would therefore be interesting to compare the real interest rates between the developed countries and the less developed countries that have low interest rates. To illustrate this, the prevailing nominal rate on postal savings in India, the United Kingdom, and Japan and the savings deposit rate in France and Malaysia is deflated by the annual percentage change in consumer prices in the respective countries, using data from the Fund publication, International Financial Statistics (December 1963 = 100 for all the countries except Malaysia, in which 1959 = 100). The resultant figures (Table 3) show that the real rate on savings deposits was negative in the developed countries in all the years and in India during 1965–67, whereas in Malaysia it was positive in all years except 1967. Although it is difficult to draw any clear-cut inferences from these data,²⁴ they do suggest that the gap between the real rates in the developed and less developed countries is perhaps not perceptibly greater than that in the nominal rates.

Table 3.

Selected Countries: Estimated Real Rate of Interest on Savings Deposits, 1965–69

(In per cent per annum)

Sources: Estimated from Table 2, national sources, and price indices given in International Monetary Fund, International Financial Statistics.

Table 3.

Selected Countries: Estimated Real Rate of Interest on Savings Deposits, 1965–69

(In per cent per annum)

Country	1965	1966	1967	1968	1969
India	−1.20	−10.00	−4.60	4.80	−0.30
Japan	−1.40	−1.80	−3.20	−0.40	−3.30
Malaysia	2.00	2.80	−1.20	3.30	4.50
United Kingdom	−1.40	−0.30	−0.20	−1.00	−2.30

Sources: Estimated from Table 2, national sources, and price indices given in International Monetary Fund, International Financial Statistics.

View Table

Table 3.

Selected Countries: Estimated Real Rate of Interest on Savings Deposits, 1965–69

(In per cent per annum)

Country	1965	1966	1967	1968	1969
India	−1.20	−10.00	−4.60	4.80	−0.30
Japan	−1.40	−1.80	−3.20	−0.40	−3.30
Malaysia	2.00	2.80	−1.20	3.30	4.50
United Kingdom	−1.40	−0.30	−0.20	−1.00	−2.30

Sources: Estimated from Table 2, national sources, and price indices given in International Monetary Fund, International Financial Statistics.

There is also an interesting parallelism between the similarity of interest rates in the organized sector of the less developed countries and the developed countries, on the one hand, and the relative average capital/output ratios (Table 4) and the average rate of return on capital in manufacturing (Table 5), on the other hand. This suggests that as the organized sectors in the less developed economies are developed enclaves they have more in common with the developed economies than with the unorganized sector in their own countries. Borrowing by the organized sector in the less developed countries is likewise from the same sector and hence it is feasible at the lower rates prevailing in that sector. Neither the governments nor the financial institutions borrow directly from the unorganized sector where interest rates are higher. Thus, the dualism of the financial sector and the coexistence of non-competing capital markets conceal the fact that the borrowing is from the low interest rate sector, which does not necessarily reflect the overall scarcity of capital in the economy.

Table 4.

Selected Countries: Average Capital/Output Ratios, c. 1960

(In per cent per annum)

Source: Raymond W. Goldsmith, Financial Structure and Development (Yale University Press, 1969), Table 6-3, p. 294.

Table 4.

Selected Countries: Average Capital/Output Ratios, c. 1960

(In per cent per annum)

Country	Including Land	Excluding Land
France	3.9	3.1
Germany	3.6	2.9
India	3.6	2.3
Israel	3.2	2.6
Japan	2.8	2.0
Mexico	3.8	2.9
United Kingdom	3.2	…
United States	4.3	3.6

Source: Raymond W. Goldsmith, Financial Structure and Development (Yale University Press, 1969), Table 6-3, p. 294.

View Table

Table 4.

Selected Countries: Average Capital/Output Ratios, c. 1960

(In per cent per annum)

Country	Including Land	Excluding Land
France	3.9	3.1
Germany	3.6	2.9
India	3.6	2.3
Israel	3.2	2.6
Japan	2.8	2.0
Mexico	3.8	2.9
United Kingdom	3.2	…
United States	4.3	3.6

Source: Raymond W. Goldsmith, Financial Structure and Development (Yale University Press, 1969), Table 6-3, p. 294.

Table 5.

Selected Countries: Average Rate of Return on Capital in Manufacturing,1948–58

Source: Bagicha Singh Minhas, An International Comparison of Factor Costs and Factor Use (Amsterdam, 1963), Table XVI, p. 88.

¹Percentage rate of return before corporate taxes.

²Based on census data.

³Based on data from the Reserve Bank of India.

Table 5.

Selected Countries: Average Rate of Return on Capital in Manufacturing,1948–58

Country	Years	Average Rate 1
Canada	1948-52	19.40
	1953-57	14.30
India	1948-52 2	20.69
	1953-55 and 1957 2	19.13
	1950-563	18.49
	1953-563	19.51
Japan	1953-57	17.92
United Kingdom	1949-53	16.75
	1954-58	15.79
United States	1949-53	19.75
	1954-58	15.98

Source: Bagicha Singh Minhas, An International Comparison of Factor Costs and Factor Use (Amsterdam, 1963), Table XVI, p. 88.

¹Percentage rate of return before corporate taxes.

²Based on census data.

³Based on data from the Reserve Bank of India.

View Table

Table 5.

Selected Countries: Average Rate of Return on Capital in Manufacturing,1948–58

Country	Years	Average Rate 1
Canada	1948-52	19.40
	1953-57	14.30
India	1948-52 2	20.69
	1953-55 and 1957 2	19.13
	1950-563	18.49
	1953-563	19.51
Japan	1953-57	17.92
United Kingdom	1949-53	16.75
	1954-58	15.79
United States	1949-53	19.75
	1954-58	15.98

Source: Bagicha Singh Minhas, An International Comparison of Factor Costs and Factor Use (Amsterdam, 1963), Table XVI, p. 88.

¹Percentage rate of return before corporate taxes.

²Based on census data.

³Based on data from the Reserve Bank of India.

The feasibility of government borrowing at low interest rates in the less developed countries reflects, first, the existence of a captive or guaranteed minimum market for government loans arising from the statutory requirements stipulating minimum ratios of investments by financial institutions and social security funds in government securities.²⁵ Such requirements coupled with the paucity of equally safe alternative investment outlets make the market for government securities less competitive in the less developed countries. Consequently, it is arguable that institutional and other investors would hold a substantial portion of their portfolios in government securities even in the absence of statutory regulations. For instance, Indian experience shows that, even before the imposition of statutory investment ratios, banks and insurers held government securities at or even above the same level that came to be required by statute. It is also not uncommon for financial intermediaries to hold “excess” (i.e., above the statutory minimum) portfolios of gilt-edged stock. Second, the low rates on government loans are possible because of the strong semimonopsonistic position of governments as borrowers in the less developed countries, which stems from the sheer size of their borrowing operations, and the imperfections of the capital market. Last, the fact that government obligations in all economies, developed or less developed, are virtually risk free makes their yield correspondingly low and uniform by eliminating the risk premium altogether.

Thus, the capacity of the less developed countries to maintain artificially low interest rates below the real cost of capital controverts the belief that the “floor interest rate” is higher than in developed countries.²⁶ This in turn is buttressed by the conventional wisdom that favors comparatively low and stable average interest rates in some of the less developed countries, reflecting the cumulative effect of various factors—intellectual, historical, and institutional. Although it is difficult to indicate their relative importance or historical sequence, it is possible to identify some of the more important influences.

A major obstacle to the adoption of bolder interest rate policies is the notion that a historical level of interest rates—say, in the range of 3-5 per cent—is in some sense a “normal” level.²⁷ The notion of a normal rate, however, stems from historical experience in the United Kingdom and elsewhere as well as being a survival from the era of war finance when it was both necessary and desirable to finance the war effort on the basis of a given and stable basic rate, such as 3 per cent. This experience appears to have strongly colored the thinking in some Asian countries, where one of the implicit assumptions of policy seems to be the desirability of financing long-term development programs on the basis of low and stable rates. But the notion that a low stable rate is normal becomes increasingly inappropriate in a developing economy, where relative prices, including the interest rate, must necessarily reflect relative scarcities of factors and goods.

The historical experience²⁸ of interest rates in the organized sector in the less developed countries has been one of variations within a comparatively narrow range of about 3–6 per cent, which has probably conditioned the authorities to low interest rates. It may be that within this range interest rates have had comparatively negligible effects. But really large changes in the rate of interest—say, 10–20 per cent—might well have a significant effect on decisions about savings and investments. Also, the fact that prevailing levels of interest in the unorganized sector are unconscionably high (20–50 per cent) creates a bias in favor of low and stable interest rates. There is an element of paradox in this attitude insofar as the acceptance of low rates in the organized sector goes together with tolerance of abnormally high rates in the unorganized sector. Logically the case is, if anything, for higher rates in the organized sector and much lower ones than those prevailing in the unorganized rural sector.

To sum up, while one may accept the general validity of the Myrdal thesis in the context of low interest rate countries in Asia, it must be subject to appropriate qualifications.

First, the choice is not between low and high rates but between rigid and flexible policies, somewhat along the lines of the Radcliffe Committee’s “three-gears” approach, with rate changes corresponding to low, medium, and high levels. The rate changes can then take the form of fractional variations (fine tuning) or else of larger steps of 1 per cent or more. But in times of high or rapid inflation the three-gears approach may be rendered inoperative, and the situation may warrant dramatic “quantum” jumps in nominal rates to offset the declines in real interest rates.

Second, even with very high rates it would not be possible to dispense wholly with discretionary credit controls because of the imperfections of the credit market, which give rise to qualitative credit rationing as well as the need to take account of social priorities. But higher interest rates would certainly help to reduce the reliance on discretionary controls and would thereby eliminate the major portion of the massive “disequilibrium” systems found necessary to maintain the existing low rates.

Third, the effects of higher interest rates on the choice of techniques and the degree of capital intensity may be blunted insofar as purely technological considerations determine the minimum size of plant and equipment in industry and construction. A more salutary effect, however, may be a more rational allocation of capital between the public and private sectors, provided that the public sector borrows its requirements in a competitive market.

Fourth, higher interest rates by themselves may not be sufficiently strong inducements either to drastically alter existing asset preferences, e.g., for land and gold, in less developed countries ²⁹ without a radical change in savings psychology or else to attract foreign loan capital. Moreover, the foreign exchange gap can be more appropriately bridged by an inflow of aid and equity capital than by loan capital.

Finally, higher interest rates would help to extend the organized sector of finance and to promote financial intermediation and the integration of the money and capital market. The maintenance of nominal interest rates in the organized sector below their true economic level results in a steady attrition of organized finance. Since only the rates in the organized sector are controlled, the rise in real rates of interest will be confined to the unorganized sectors, resulting in a steady diversion of savings to that sector in search of higher rates of return. This may result in an increase of either consumption or direct investment in the unorganized sector, since lenders in this sector are more prone to make consumption loans. Thus, attrition of organized finance is accompanied by a rise in consumption, a fall in investment and savings, misallocation of savings to investment, and inefficient use of real resources.

Another consequence of this process is the possible adverse impact on the monetization of the economy, at least in conditions of rising prices. Inflation functions as a kind of tax on cash balances and leads to a steady decline in the holdings of money and also in financial assets. On the other hand, the pace of monetization could not be said to depend on the level of interest rates.³⁰

The savings pattern in Asia

The potentialities of an active interest rate policy depend on the extent to which the voluntary financial savings of the household sector are responsive to variations in the level and structure of interest rates. The available data on the savings pattern of some selected Asian countries emphasize, first, the sizable share of the household sector in gross savings (Table 6); second, the relatively small proportion of financial assets—between 35 and 45 per cent—except for Japan, 82 per cent, and Ceylon, nearly 66 per cent (Table 7); and, finally, the overwhelming predominance—more than 70 per cent—of the voluntary component in household savings and the correspondingly negligible role of compulsory and contractual savings (Table 8). Together these factors suggest a favorable environment for the use of interest rates to stimulate voluntary financial saving by households. In particular, the still negligible role of contractual and compulsory savings, unlike the situation in the developed countries, indicates a much wider potential scope for interest-sensitive savings by households.

Table 6.

Selected Asian Countries: Sectoral Distribution of Gross Savings, 1954–59

(In per cent)

Source: United Nations, Economic Commission for Asia and the Far East, Economic Bulletin for Asia and the Far East, December 1962, p. 4.

¹Derived as a residual by deducting identifiable savings in other sectors.

²Net savings.

Table 6.

Selected Asian Countries: Sectoral Distribution of Gross Savings, 1954–59

(In per cent)

Country		Government	Corporations	Households
Ceylon
	1955	50.8	6.7	42.5
	1956	50.9	14.3	34.8
	1957	50.7	18.0	31.4
	1958	10.2	18.3	71.5
	1959	7.8	18.3	73.9
	1955-59	38.4	13.8	47.8
China
	1958	41.6	4.4	54.0 1
	1959	33.6	7.6	58.8 1
	1958-59	37.1	6.2	56.7 1
India 2
	1954-55	5.6	7.2	87.2
	1955-56	3.3	7.9	88.8
	1956-57	9.4	7.2	84.8
	1957-58	13.2	2.1	84.7
	1955-58	8.0	5.7	86.3
Korea
	1958	–83.3	48.0	135.3
	1959	–36.0	22.0	114.0
	1958-59	–51.6	30.6	121.0

Source: United Nations, Economic Commission for Asia and the Far East, Economic Bulletin for Asia and the Far East, December 1962, p. 4.

¹Derived as a residual by deducting identifiable savings in other sectors.

²Net savings.

View Table

Table 6.

Selected Asian Countries: Sectoral Distribution of Gross Savings, 1954–59

(In per cent)

Country		Government	Corporations	Households
Ceylon
	1955	50.8	6.7	42.5
	1956	50.9	14.3	34.8
	1957	50.7	18.0	31.4
	1958	10.2	18.3	71.5
	1959	7.8	18.3	73.9
	1955-59	38.4	13.8	47.8
China
	1958	41.6	4.4	54.0 1
	1959	33.6	7.6	58.8 1
	1958-59	37.1	6.2	56.7 1
India 2
	1954-55	5.6	7.2	87.2
	1955-56	3.3	7.9	88.8
	1956-57	9.4	7.2	84.8
	1957-58	13.2	2.1	84.7
	1955-58	8.0	5.7	86.3
Korea
	1958	–83.3	48.0	135.3
	1959	–36.0	22.0	114.0
	1958-59	–51.6	30.6	121.0

Source: United Nations, Economic Commission for Asia and the Far East, Economic Bulletin for Asia and the Far East, December 1962, p. 4.

¹Derived as a residual by deducting identifiable savings in other sectors.

²Net savings.

Table 7.

Selected Asian Countries: Main Forms of Gross Household Savings, 1955–59

(In per cent)

Source: United Nations, Economic Commission for Asia and the Far East, Economic Bulletin for Asia and the Far East, December 1962, p. 6.

¹Net savings basis, i.e., net of capital consumption allowances.

²Total assets.

Table 7.

Selected Asian Countries: Main Forms of Gross Household Savings, 1955–59

(In per cent)

	Financial Assets
		Deposits				Contractual savings
Country and Period	Total	Total	Currency	Demand deposits	Savings and time deposits	Life insurance	Other	Securities	Other	Tangible Assets	Liabilities
Ceylon, 1955-59	65.6	29.6	13.3	0.7	15.6	20.6	8.0	5.3	2.1	77.9	43.6
China, 1958-59	35.9	—	—	—	—	—	—	—	—	64.0	—
India,1 1955–58	45.1	16.3	13.6	2.7 2		3.1	10.1	12.7	2.9	54.8	—
Japan, 1955-58	82.0	56.2	3.0	–0.03	53.2	10.3	0.2	8.4	6.9	41.6	23.7
Korea, 1958-59	46.1	35.0	12.6	9.1	13.3	0.6	0.6	13.3	–3.6	90.9	37.0
Malaysia, 1955-58	39.2	2.2	5.1	-2.9 2		7.6	19.4	10.0	—	60.7	—
Philippines, 1955-59	42.5	19.9	5.6	5.0	9.3	14.7	1.9	13.8	—	63.1	5.5

Source: United Nations, Economic Commission for Asia and the Far East, Economic Bulletin for Asia and the Far East, December 1962, p. 6.

¹Net savings basis, i.e., net of capital consumption allowances.

²Total assets.

View Table

Table 7.

Selected Asian Countries: Main Forms of Gross Household Savings, 1955–59

(In per cent)

	Financial Assets
		Deposits				Contractual savings
Country and Period	Total	Total	Currency	Demand deposits	Savings and time deposits	Life insurance	Other	Securities	Other	Tangible Assets	Liabilities
Ceylon, 1955-59	65.6	29.6	13.3	0.7	15.6	20.6	8.0	5.3	2.1	77.9	43.6
China, 1958-59	35.9	—	—	—	—	—	—	—	—	64.0	—
India,1 1955–58	45.1	16.3	13.6	2.7 2		3.1	10.1	12.7	2.9	54.8	—
Japan, 1955-58	82.0	56.2	3.0	–0.03	53.2	10.3	0.2	8.4	6.9	41.6	23.7
Korea, 1958-59	46.1	35.0	12.6	9.1	13.3	0.6	0.6	13.3	–3.6	90.9	37.0
Malaysia, 1955-58	39.2	2.2	5.1	-2.9 2		7.6	19.4	10.0	—	60.7	—
Philippines, 1955-59	42.5	19.9	5.6	5.0	9.3	14.7	1.9	13.8	—	63.1	5.5

Source: United Nations, Economic Commission for Asia and the Far East, Economic Bulletin for Asia and the Far East, December 1962, p. 6.

¹Net savings basis, i.e., net of capital consumption allowances.

²Total assets.

Table 8.

Selected Asian Countries: Compulsory, Contractual, and Voluntary Savings of Households, on Net Savings Basis, 1955–59

(In per cent)

Source: United Nations, Economic Commission for Asia and the Far East, Economic Bulletin for Asia and the Far East, December 1962, p. 10.

¹1955–59.

²1955–58.

³1958–59.

Table 8.

Selected Asian Countries: Compulsory, Contractual, and Voluntary Savings of Households, on Net Savings Basis, 1955–59

(In per cent)

		1955	1956	1957	1958	1959	Total
Ceylon
	Compulsory	4.2	5.4	7.7	0.4	20.7	8.0 1
	Contractual	11.5	20.8	42.0	23.5	20.9	21.2 1
	Voluntary	84.3	73.8	50.3	76.1	58.4	70.81
							100.0
India
	Compulsory	9.0	9.3	12.0	10.6	—	10.1 2
	Contractual	1.8	4.3	2.1	4.0	—	3.1 2
	Voluntary	89.2	86.4	85.8	85.4	—	86.82
							100.0
Japan
	Compulsory	—	—	—	—	—	—
	Contractual	7.1	9.8	10.1	14.4	—	10.3 2
	Voluntary	92.9	90.2	89.9	85.6	—	89.72
							100.0
Korea
	Compulsory	—	—	—	—	—	—
	Contractual	—	—	—	0.8	2.8	2.2 3
	Voluntary	—	—	—	99.2	97.2	97.83
							100.0
Malaysia
	Compulsory	13.2	16.5	42.4	19.5	—	19.4 2
	Contractual	4.8	6.1	17.2	8.4	—	7.7 2
	Voluntary	82.0	77.4	40.4	72.1	—	72.92
							100.0
Philippines
	Compulsory	8.4	6.5	7.2	10.1	11.3	8.8 1
	Contractual	7.9	6.0	5.6	7.6	8.4	7.1 1
	Voluntary	83.7	87.5	87.2	82.3	80.3	84.11
							100.0

Source: United Nations, Economic Commission for Asia and the Far East, Economic Bulletin for Asia and the Far East, December 1962, p. 10.

¹1955–59.

²1955–58.

³1958–59.

View Table

Table 8.

Selected Asian Countries: Compulsory, Contractual, and Voluntary Savings of Households, on Net Savings Basis, 1955–59

(In per cent)

		1955	1956	1957	1958	1959	Total
Ceylon
	Compulsory	4.2	5.4	7.7	0.4	20.7	8.0 1
	Contractual	11.5	20.8	42.0	23.5	20.9	21.2 1
	Voluntary	84.3	73.8	50.3	76.1	58.4	70.81
							100.0
India
	Compulsory	9.0	9.3	12.0	10.6	—	10.1 2
	Contractual	1.8	4.3	2.1	4.0	—	3.1 2
	Voluntary	89.2	86.4	85.8	85.4	—	86.82
							100.0
Japan
	Compulsory	—	—	—	—	—	—
	Contractual	7.1	9.8	10.1	14.4	—	10.3 2
	Voluntary	92.9	90.2	89.9	85.6	—	89.72
							100.0
Korea
	Compulsory	—	—	—	—	—	—
	Contractual	—	—	—	0.8	2.8	2.2 3
	Voluntary	—	—	—	99.2	97.2	97.83
							100.0
Malaysia
	Compulsory	13.2	16.5	42.4	19.5	—	19.4 2
	Contractual	4.8	6.1	17.2	8.4	—	7.7 2
	Voluntary	82.0	77.4	40.4	72.1	—	72.92
							100.0
Philippines
	Compulsory	8.4	6.5	7.2	10.1	11.3	8.8 1
	Contractual	7.9	6.0	5.6	7.6	8.4	7.1 1
	Voluntary	83.7	87.5	87.2	82.3	80.3	84.11
							100.0

Source: United Nations, Economic Commission for Asia and the Far East, Economic Bulletin for Asia and the Far East, December 1962, p. 10.

¹1955–59.

²1955–58.

³1958–59.

This leads to the more difficult question of interest elasticity of personal savings in Asia, on which the available evidence is inconclusive and sometimes conflicting. According to a recent comparative evaluation by J. G. Williamson of some of the major determinants of personal savings in Asia,³¹ combining both temporal analysis of individual Asian nations and intertemporal cross-section analysis of a large group of Asian countries, “higher interest rates are associated, if anything, with lower real saving in Asia. The explanation would appear to lie in the fact that the savings and investment decisions are highly interdependent in the Asian household sector …. interest rates appear to influence the short-run savings decision far more powerfully than the long-run savings decision.” ³² One of the findings of the study was that for Asia the net impact of real interest rate movements on personal saving was either negative or insignificant (see Table 9).

Table 9.

Selected Asian Countries: Real Interest Rates and Personal Savings¹, ²

Source: Jeffrey G. Williamson, “Personal Saving in Developing Nations: An Intertemporal Cross-Section from Asia,” The Economic Record, June 1968, Table VI, p. 208.

¹Real interest rates (r_{i, t}) as deflated by cost of living index; personal savings as dependent variable (S_{i, t}); normal disposable income (${\hat{Y}}_{i,t}^d$);and transitory income (${\stackrel{*}{Y}}_{i,t}^d$).

²* indicates lack of significance at 90 per cent level; standard errors are shown in parentheses.

Table 9.

Selected Asian Countries: Real Interest Rates and Personal Savings¹, ²

		Regression Coefficients of
Country	Intercept	$r_{i,t}-\frac{{\mathrm{\Delta }P}_{i,t}}{P_{i,t-1}}$	${\hat{Y}}_{i,t}^d$	${\stackrel{*}{Y}}_{i,t}^d$	${\overline{R}}^2$
Burma	0.3859*	0.0253*	0.1002*	0.3621	0.369
	(4.7628)	(0.1060)	(0.0965)	(0.2056)
China	−21.9735	−0.0674*	0.3002	−0.1465*	0.672
	(9.1466)	(0.0865)	(0.0963)	(0.4268)
India	17.6000*	−0.0392*	−0.1570*	0.0823*	0.129
	(38.7668)	(0.2365)	(0.6024)	(0.4474)
Japan	−27.2204	−0.7612	0.3019	0.6372	0.967
	(4.5385)	(0.2543)	(0.0188)	(0.2486)
Korea	0.9218*	−0.0018*	−0.0685*	1.0950*	0.343
	(1.6962)	(0.0166)	(0.1403)	(0.8783)
Philippines	−24.4289	−0.3027	0.2581	0.8505	0.753
	(8.1308)	(0.1839)	(0.0815)	(0.4271)

Source: Jeffrey G. Williamson, “Personal Saving in Developing Nations: An Intertemporal Cross-Section from Asia,” The Economic Record, June 1968, Table VI, p. 208.

¹Real interest rates (r_{i, t}) as deflated by cost of living index; personal savings as dependent variable (S_{i, t}); normal disposable income (${\hat{Y}}_{i,t}^d$);and transitory income (${\stackrel{*}{Y}}_{i,t}^d$).

²* indicates lack of significance at 90 per cent level; standard errors are shown in parentheses.

View Table

Table 9.

Selected Asian Countries: Real Interest Rates and Personal Savings¹, ²

		Regression Coefficients of
Country	Intercept	$r_{i,t}-\frac{{\mathrm{\Delta }P}_{i,t}}{P_{i,t-1}}$	${\hat{Y}}_{i,t}^d$	${\stackrel{*}{Y}}_{i,t}^d$	${\overline{R}}^2$
Burma	0.3859*	0.0253*	0.1002*	0.3621	0.369
	(4.7628)	(0.1060)	(0.0965)	(0.2056)
China	−21.9735	−0.0674*	0.3002	−0.1465*	0.672
	(9.1466)	(0.0865)	(0.0963)	(0.4268)
India	17.6000*	−0.0392*	−0.1570*	0.0823*	0.129
	(38.7668)	(0.2365)	(0.6024)	(0.4474)
Japan	−27.2204	−0.7612	0.3019	0.6372	0.967
	(4.5385)	(0.2543)	(0.0188)	(0.2486)
Korea	0.9218*	−0.0018*	−0.0685*	1.0950*	0.343
	(1.6962)	(0.0166)	(0.1403)	(0.8783)
Philippines	−24.4289	−0.3027	0.2581	0.8505	0.753
	(8.1308)	(0.1839)	(0.0815)	(0.4271)

Source: Jeffrey G. Williamson, “Personal Saving in Developing Nations: An Intertemporal Cross-Section from Asia,” The Economic Record, June 1968, Table VI, p. 208.

¹Real interest rates (r_{i, t}) as deflated by cost of living index; personal savings as dependent variable (S_{i, t}); normal disposable income (${\hat{Y}}_{i,t}^d$);and transitory income (${\stackrel{*}{Y}}_{i,t}^d$).

²* indicates lack of significance at 90 per cent level; standard errors are shown in parentheses.

However, another regression analysis, by K. L. Gupta,³³ using the same variables as Williamson’s but a different and more reliable set of primary data on savings for India (estimates by the Reserve Bank of India instead of the National Council of Applied Economic Research) arrived at results just the opposite of Williamson’s. Gupta’s results (Table 10) show that while the real rate of interest is not significant at the aggregate level it is more influential in determining personal savings at the per capita level. Apart from the use of more reliable primary data, Gupta’s analysis, unlike Williamson’s, which relies on a single rate of interest as the index of return on financial assets, considers the following real rates of interest as alternatives.

Table 10.

India: Interest Rates and Personal Savings¹

Source: K.L. Gupta, “Personal Saving in Developing Nations: Further Evidence,” The Economic Record, Vol. 46 (1970), p. 248.

¹Standard errors are shown in parentheses.

Table 10.

India: Interest Rates and Personal Savings¹

Aggregate
$S=-982.7+\underset{\left(0.0310\right)}{0.15019}{Y}_d+\underset{\left(118.280\right)}{25.965}{r}_s$	R = 0.849
$S=-1008.6+0.12875Y_d^{\prime }-0.32175Y_d+112.06r_s$	R = 0.871
Per capita
$S=-94.386+\underset{\left(0.1327\right)}{0.37249}{Y}_d+\underset{\left(2.0322\right)}{3.7892}{r}_{Ig}$	R = 0.698
$S=-94.979+\underset{\left(0.1375\right)}{0.3859}{Y}_d+\underset{\left(2.9825\right)}{4.9280}{r}_s$	R = 0.677
$S=-46.182+\underset{\left(0.2757\right)}{0.1660}{Y}_d^{\prime }-\underset{\left(0.1818\right)}{0.4774}{Y}_d+\underset{\left(2.501\right)}{5.009}{r}_{Ig}$	R = 0.723
$S=-96.442+\underset{\left(0.2613\right)}{0.39178}{Y}_d^{\prime }-\underset{\left(0.1788\right)}{0.38311}{Y}_d+\underset{\left(3.1846\right)}{4.912}{r}_s$	R = 0.678

Source: K.L. Gupta, “Personal Saving in Developing Nations: Further Evidence,” The Economic Record, Vol. 46 (1970), p. 248.

¹Standard errors are shown in parentheses.

View Table

Table 10.

India: Interest Rates and Personal Savings¹

Aggregate
$S=-982.7+\underset{\left(0.0310\right)}{0.15019}{Y}_d+\underset{\left(118.280\right)}{25.965}{r}_s$	R = 0.849
$S=-1008.6+0.12875Y_d^{\prime }-0.32175Y_d+112.06r_s$	R = 0.871
Per capita
$S=-94.386+\underset{\left(0.1327\right)}{0.37249}{Y}_d+\underset{\left(2.0322\right)}{3.7892}{r}_{Ig}$	R = 0.698
$S=-94.979+\underset{\left(0.1375\right)}{0.3859}{Y}_d+\underset{\left(2.9825\right)}{4.9280}{r}_s$	R = 0.677
$S=-46.182+\underset{\left(0.2757\right)}{0.1660}{Y}_d^{\prime }-\underset{\left(0.1818\right)}{0.4774}{Y}_d+\underset{\left(2.501\right)}{5.009}{r}_{Ig}$	R = 0.723
$S=-96.442+\underset{\left(0.2613\right)}{0.39178}{Y}_d^{\prime }-\underset{\left(0.1788\right)}{0.38311}{Y}_d+\underset{\left(3.1846\right)}{4.912}{r}_s$	R = 0.678

Source: K.L. Gupta, “Personal Saving in Developing Nations: Further Evidence,” The Economic Record, Vol. 46 (1970), p. 248.

¹Standard errors are shown in parentheses.

(a) rr	=	short-term treasury bills
(b) rIg	=	long-term government bonds
(c) rIp	=	private securities
(d) rt	=	commercial bank time deposits
(e) rg	=	nonbank savings deposits

View Table

(a) rr	=	short-term treasury bills
(b) rIg	=	long-term government bonds
(c) rIp	=	private securities
(d) rt	=	commercial bank time deposits
(e) rg	=	nonbank savings deposits

The results of Gupta’s analysis (Table 10) for India show that in all the equations the coefficient has a positive sign, suggesting that higher real interest rates lead to higher real savings.

Although the econometric evidence on the interest sensitivity of personal savings in Asia is thus inconclusive and conflicting, it points to the potentially greater role for interest rates in a country like India.³⁴ To that extent, it provides a promising statistical foundation for the Myrdal thesis. But more than such econometric investigations, with all their inherent limitations, a review of the experience of such countries as China and Korea might yield more meaningful inferences regarding the scope and limitations of an active interest rate policy for developing economies.

Experience of high interest rate strategies in Asia

China

Chinese experience since 1949 is of exceptional significance in highlighting the potentialities of a conscious and purposive interest rate policy in an economy subject to the strains of civil war and postwar inflation as well as developmental expenditures. China is rightly regarded as a pioneer and leading exponent of a high but flexible interest rate strategy.

The monetary history of China since 1949 affords a classic example of conditions in which interest rates become one of the major anti-inflationary instruments. The extent of the inflation that had to be combated and its repercussions on the level of interest rates are brought out in Table 11.

Table 11.

China: Prices and Interest Rates, 1940–46

(In per cent per annum)

Source: United Nations, Department of Economic Affairs, Inflationary and Deflationary Tendencies, 1946–1948 (Sales No.: 1949.II.A.1), p. 48.

Table 11.

China: Prices and Interest Rates, 1940–46

(In per cent per annum)

Year	Rate of Increase of Wholesale Prices	Black Market Rate of Interest
1940	158	22
1941	177	32
1942	180	33
1943	202	136
1944	224	197
1945	253	270
1946	151	185

Source: United Nations, Department of Economic Affairs, Inflationary and Deflationary Tendencies, 1946–1948 (Sales No.: 1949.II.A.1), p. 48.

View Table

Table 11.

China: Prices and Interest Rates, 1940–46

(In per cent per annum)

Year	Rate of Increase of Wholesale Prices	Black Market Rate of Interest
1940	158	22
1941	177	32
1942	180	33
1943	202	136
1944	224	197
1945	253	270
1946	151	185

Source: United Nations, Department of Economic Affairs, Inflationary and Deflationary Tendencies, 1946–1948 (Sales No.: 1949.II.A.1), p. 48.

But inflationary pressures and rising prices showed no signs of abatement even after the consolidation of the regime in China after 1945 and, if anything, were aggravated by the growing volume of military expenditure and the influx of refugees from the mainland. In this situation, apart from the political difficulties of higher taxation and compulsory borrowing, there was not much hope of evoking any public response to government bonds or of increased accruals to savings bank accounts despite an annual average rate of increase of 85 per cent in wholesale prices.³⁵ Under these conditions, the Chinese authorities were inevitably led to rely on economic incentives rather than the purely patriotic impulses of the populace. In the monetary sphere this pointed to the use of sufficiently attractive interest rates to induce the public to save more.

But considering the prevailing rate of price rise and the fact that interest rates were already high, evidenced among other things by the payment of interest even on current account deposits (3.24 per cent per annum in 1949), it was evident that the authorities would really need to pitch the rate on new savings instruments at extraordinarily high levels. The Chinese authorities boldly broke through the psychological barrier, which so often inhibits the raising of nominal interest rates above conventional levels, by introducing in March 1950 a special system of preferential deposits of one-month, two-month, and three-month maturities carrying a rate of 7 per cent a month (i.e., 125 per cent per annum compounded monthly). The choice of very short maturities was significant, inasmuch as with the high liquidity preference prevailing under conditions of hyperinflation, it was unrealistic to expect the public to buy bonds of long-term or even medium-term maturities. Thus, initially, the policy comprised a combination of very high interest rates on very short-term maturities. But there was no irrevocable commitment to a policy of rising rates, since this would have undermined the market by inducing investors to hold back in anticipation of further rises. Encouraged both by the response to the new deposits as well as by the decline in the wholesale price index between May and July 1950, the authorities reduced the rate of interest on one-month deposits in July 1950 by one half, to 3½ per cent a month. This reduction led to a slight fall in time deposits, but even so this reaction of depositors proved to the authorities that they had been able to arrive at the “critical area” in which monetary savings had become responsive to interest rate changes. Nevertheless, the authorities reduced the rate to 3.0 per cent a month on one-month deposits. The cut coupled with the resumption of a sharp rise in prices may have led to the fall in the preferential interest rate deposits to a very modest level of NT$21 million in January 1951.

The authorities were concerned, however, to ensure that the rise in the deposit rates was not communicated to loan rates, since this would have affected the working capital requirements of trade and industry. Accordingly, the strategy was adapted to ensure a ceiling on loan rates as well as a “floor” for deposit rates. But the progressive raising of rates on deposits without a corresponding increase in loan rates created an “inverted” interest rate structure with a negative differential between the preferential deposit and loan rates, since the former were raised to higher levels than the latter. To meet this situation a “redeposit facility” was created whereby banks were given the option of placing excess deposits (i.e., those for which they had no immediate outlet by way of investment or loans) in the Bank of Taiwan ³⁶ at rates equal to or above those paid by the commercial banks. This, in effect, protected the commercial banks against losses on the preferential deposits. This facility has proved quite popular with the banks, especially since 1951, as evidenced by the rise of redeposits from NT$6 million in March 1951 to NT$329 million in August 1952. The spectacular rise in redeposits reflected the weak demand for credit relative to the growth of bank deposits in response to the high interest rates. The loss suffered by the Bank of Taiwan in paying higher interest on redeposits than its earnings from loans and investments was substantially offset by the earnings from U. S. counterpart funds deposited with it. But, more importantly, the net cost of “redeposits” to the Bank of Taiwan in financial terms was more than offset by its efficacy in combating inflation, an apt example of external economies of the high interest rate policy.

Thereafter the rate was again increased to 4.2 per cent a month (64 per cent per annum) and was held steady at that level for the next year. By April 1953 the volume of preferential deposits had risen to NT$350 million. At this stage the authorities, reacting to the restoration of public confidence, attempted to stretch the maturity of deposits while keeping interest rates constant through the introduction of six-month deposit certificates at 4.2 per cent a month and one-year deposit certificates at a rate of 3.0 per cent a month. During the second phase of this policy the authorities, encouraged by the continued rise of deposits, reduced rates from 2 per cent to 1.2 per cent (on one-month deposits), 2.15 per cent to 1.3 per cent (three-month deposits), 2.3 per cent to 1.5 per cent (six-month deposits), and 3 per cent to 2 per cent (one-year deposits). The result was a switch to longer maturities, and by the end of 1953 the six-month and one-year deposits accounted for almost half the total deposits.

But despite the subsequent growth of preferential time deposits, there were apprehensions about the excessive potential liquidity and its possible activation through the use of these deposits as collateral for bank credit, a possibility that exposed a gap in the anti-inflationary policy. Consequently, rate adjustments in March 1956 were designed to render shorter maturities less attractive by cuts of 15 per cent and 9 per cent on one-month and three-month deposits, respectively. These reductions were coupled with the offer of premiums of 15.4 per cent and 12.5 per cent on six-month and one-year deposits, respectively, if the depositor chose to forgo borrowing privileges against these deposits. But even this did not remove the inflationary potential of increased liquidity, and therefore all time deposits were declared ineligible as bank collateral. A substantial switch to longer maturities followed, and by the end of 1956 a little more than one fourth of the time deposits were for one year. But the fact that 1956 was also the first year in which preferential time deposits failed to register an annual increase implied that for a large number of depositors the increased premium did not sufficiently compensate for the reduced liquidity. This demonstrated that depositors were as conscious of liquidity as of yield. However, a significant development in 1957 was the abolition of interest on checking accounts with banks (paid at the rate of 3.24 per cent in 1949, 1.62 per cent in 1950-55, and 0.90 per cent in 1956), which betokened the restoration of financial normalcy in the economy.

In the second half of 1957 two-year preferential deposits were introduced, carrying a rate of 1.8 per cent a month, the rate previously paid on one-year deposits. Concurrently, the rate on one-year deposits was reduced to 1.35 per cent monthly and the one-month deposits were discontinued. The three-month deposits (0.85 per cent a month) lost their popular appeal. The volume of one-year deposits more than doubled in 1958; the six-month deposits too approximately doubled between mid-1957 and the end of 1958. By the end of 1958 preferential deposits amounted to NT$ 1,508 million (29 per cent of net money supply), of which almost half was in one-year deposits (1.65 per cent a month) and one third in six-month deposits (1.35 per cent a month). The remainder was accounted for by three-month deposits (0.85 per cent) and two-year deposits (1.8 per cent). At the end of 1958 all preferential interest rate time deposits were terminated in favor of regular time deposits.

In 1959 the high interest rate strategy was extended to government borrowing with the issue of bonds with maturities of 12 to 30 months at rates varying from 9 per cent to 18 per cent. The interest on the 18 per cent bonds, which were sold exclusively to nonbank investors, was paid quarterly and was also exempt from income tax. At the same time the authorities encouraged commercial firms to raise capital through bonds carrying attractive rates of return. During 1958-59 approximately three fourths of private two-year bond issues at 20 per cent per annum were guaranteed by the Government. The nonguaranteed two-year bonds carried an annual rate of 26.4 per cent. At that time secured and unsecured bank loans were being made at 19.8 per cent and 22.3 per cent per annum, respectively, while private moneylenders charged from 36 per cent to 40 per cent.

The years 1958-59, in retrospect, marked the completion of the phase of stabilization that created an environment wherein interest rates could be viewed as a normal instrument of policy. Since that time the authorities have steadily endeavored to reduce nominal rates on deposits and loans in keeping with the improvement in the overall economic situation, the increase in quasi-money, and the relative stability of the cost of living (Tables 12 and 13).

Table 12.

China: Interest Rates on Deposits; Quasi-Money; and Private Savings, 1949–69

(Rates in per cent per annum; ¹ amounts in billions of new Taiwan dollars)

Sources: Central Bank of China, Taiwan Financial Statistics Monthly; Directorate-General of Budgets, Accounts, and Statistics, National Income of the Republic of China.

¹Rates for savings deposits, discounts, call loans, and time loans are monthly rates at the end of December converted to annual rates. Real rates are adjusted for changes in wholesale prices (A) and consumer prices (B); the rates for 1968 are adjusted for price changes between December 1967 and October 1968.

²Rates for 1949-56 are applicable only to banks other than the Bank of Taiwan. The rate on Bank of Taiwan demand (checking) deposits was 0.90 from June 1950 to June 1957, when interest was abolished for the Bank of Taiwan. Interest was abolished for all other banks in the following month. The rate on Bank of Taiwan demand (passbook) deposits was 3.6 per cent per annum from June 1950 to July 1957; from that date until 1961 the rate was 2.88 per cent per annum, the same as paid by other banks.

³Rate was applicable only to banks other than the Bank of Taiwan until January 1959, when the latter started to accept three-month deposits at the same rate, 9.0 per cent per annum.

⁴From 1953 to 1958 this rate, applicable to all banks, refers to one-year preferential deposits, which from April 1953 could be pledged as collateral on mortgage loans; after that date the privilege was withdrawn. From 1959, the rate refers to one-year savings deposits, which were introduced in January of that year. Since then no new preferential deposits have been accepted. From July 1, 1963 the same rate has been applied to two-year and three-year deposits.

⁵From International Monetary Fund, International Financial Statistics.

Table 12.

China: Interest Rates on Deposits; Quasi-Money; and Private Savings, 1949–69

(Rates in per cent per annum; ¹ amounts in billions of new Taiwan dollars)

	Demand Deposits 2			Quasi-Money
	Nominal rates			Nominal rates		Real Rates on One-Year Deposits		Amount of Quasi-Money 5	Private Savings as Percentage of Gross National Product at Current Prices
Year	Checking	Passbook	Amount	Three-month deposits 3	One-year deposits 4	(A)	B)
1949	3.24	16.20	0.09	54.00	—	—	—	—	—
1950	1.62	8.10	0.20	39.60	—	—	—	—	—
1951	1.62	8.10	0.34	54.00	—	—	—	—	—
1952	1.62	8.10	0.52	25.80	—	—	—	0.52	—
1953	1.62	5.40	0.70	15.60	24.00	8.88	2.00	0.69	3.9
1954	1.62	5.40	0.93	13.20	19.20	21.63	20.20	0.88	3.5
1955	1.62	5.40	1.15	13.20	19.20	−9.19	5.20	1.03	3.5
1956	0.90	3.60	1.57	12.00	21.60	15.97	8.60	1.07	4.0
1957	—	2.88	1.78	10.20	19.80	18.15	18.80	1.45	3.7
1958	—	2.88	2.57	10.20	19.80	13.93	17.80	2.57	4.3
1959	—	2.88	2.88	9.00	17.04	5.35	5.84	3.46	3.8
1960	—	2.88	3.30	9.00	17.04	5.33	−2.06	4.72	5.0
1961	—	1.44	4.12	7.20	14.40	14.34	10.30	7.75	7.2
1962	—	1.44	4.30	6.48	13.32	6.83	10.52	9.56	7.4
1963	—	1.44	5.76	6.00	12.00	6.50	11.30	12.47	9.5
1964	—	1.44	7.89	6.00	10.80	13.30	10.20	15.81	10.5
1965	—	1.44	9.02	6.00	10.80	12.44	10.10	18.43	10.3
1966	—	1.44	10.27	6.00	10.08	7.84	8.48	23.90	12.7
1967	—	1.44	12.99	5.40	9.72	6.95	7.02	29.57	13.2
1968	—	1.44	15.12	6.48	9.72	8.19	3.42	33.62	11.5
1969	—	1.44	17.14	6.48	9.72	8.13	4.02	41.03	…

Sources: Central Bank of China, Taiwan Financial Statistics Monthly; Directorate-General of Budgets, Accounts, and Statistics, National Income of the Republic of China.

¹Rates for savings deposits, discounts, call loans, and time loans are monthly rates at the end of December converted to annual rates. Real rates are adjusted for changes in wholesale prices (A) and consumer prices (B); the rates for 1968 are adjusted for price changes between December 1967 and October 1968.

²Rates for 1949-56 are applicable only to banks other than the Bank of Taiwan. The rate on Bank of Taiwan demand (checking) deposits was 0.90 from June 1950 to June 1957, when interest was abolished for the Bank of Taiwan. Interest was abolished for all other banks in the following month. The rate on Bank of Taiwan demand (passbook) deposits was 3.6 per cent per annum from June 1950 to July 1957; from that date until 1961 the rate was 2.88 per cent per annum, the same as paid by other banks.

³Rate was applicable only to banks other than the Bank of Taiwan until January 1959, when the latter started to accept three-month deposits at the same rate, 9.0 per cent per annum.

⁴From 1953 to 1958 this rate, applicable to all banks, refers to one-year preferential deposits, which from April 1953 could be pledged as collateral on mortgage loans; after that date the privilege was withdrawn. From 1959, the rate refers to one-year savings deposits, which were introduced in January of that year. Since then no new preferential deposits have been accepted. From July 1, 1963 the same rate has been applied to two-year and three-year deposits.

⁵From International Monetary Fund, International Financial Statistics.

View Table

Table 12.

China: Interest Rates on Deposits; Quasi-Money; and Private Savings, 1949–69

(Rates in per cent per annum; ¹ amounts in billions of new Taiwan dollars)

	Demand Deposits 2			Quasi-Money
	Nominal rates			Nominal rates		Real Rates on One-Year Deposits		Amount of Quasi-Money 5	Private Savings as Percentage of Gross National Product at Current Prices
Year	Checking	Passbook	Amount	Three-month deposits 3	One-year deposits 4	(A)	B)
1949	3.24	16.20	0.09	54.00	—	—	—	—	—
1950	1.62	8.10	0.20	39.60	—	—	—	—	—
1951	1.62	8.10	0.34	54.00	—	—	—	—	—
1952	1.62	8.10	0.52	25.80	—	—	—	0.52	—
1953	1.62	5.40	0.70	15.60	24.00	8.88	2.00	0.69	3.9
1954	1.62	5.40	0.93	13.20	19.20	21.63	20.20	0.88	3.5
1955	1.62	5.40	1.15	13.20	19.20	−9.19	5.20	1.03	3.5
1956	0.90	3.60	1.57	12.00	21.60	15.97	8.60	1.07	4.0
1957	—	2.88	1.78	10.20	19.80	18.15	18.80	1.45	3.7
1958	—	2.88	2.57	10.20	19.80	13.93	17.80	2.57	4.3
1959	—	2.88	2.88	9.00	17.04	5.35	5.84	3.46	3.8
1960	—	2.88	3.30	9.00	17.04	5.33	−2.06	4.72	5.0
1961	—	1.44	4.12	7.20	14.40	14.34	10.30	7.75	7.2
1962	—	1.44	4.30	6.48	13.32	6.83	10.52	9.56	7.4
1963	—	1.44	5.76	6.00	12.00	6.50	11.30	12.47	9.5
1964	—	1.44	7.89	6.00	10.80	13.30	10.20	15.81	10.5
1965	—	1.44	9.02	6.00	10.80	12.44	10.10	18.43	10.3
1966	—	1.44	10.27	6.00	10.08	7.84	8.48	23.90	12.7
1967	—	1.44	12.99	5.40	9.72	6.95	7.02	29.57	13.2
1968	—	1.44	15.12	6.48	9.72	8.19	3.42	33.62	11.5
1969	—	1.44	17.14	6.48	9.72	8.13	4.02	41.03	…

Sources: Central Bank of China, Taiwan Financial Statistics Monthly; Directorate-General of Budgets, Accounts, and Statistics, National Income of the Republic of China.

¹Rates for savings deposits, discounts, call loans, and time loans are monthly rates at the end of December converted to annual rates. Real rates are adjusted for changes in wholesale prices (A) and consumer prices (B); the rates for 1968 are adjusted for price changes between December 1967 and October 1968.

²Rates for 1949-56 are applicable only to banks other than the Bank of Taiwan. The rate on Bank of Taiwan demand (checking) deposits was 0.90 from June 1950 to June 1957, when interest was abolished for the Bank of Taiwan. Interest was abolished for all other banks in the following month. The rate on Bank of Taiwan demand (passbook) deposits was 3.6 per cent per annum from June 1950 to July 1957; from that date until 1961 the rate was 2.88 per cent per annum, the same as paid by other banks.

³Rate was applicable only to banks other than the Bank of Taiwan until January 1959, when the latter started to accept three-month deposits at the same rate, 9.0 per cent per annum.

⁴From 1953 to 1958 this rate, applicable to all banks, refers to one-year preferential deposits, which from April 1953 could be pledged as collateral on mortgage loans; after that date the privilege was withdrawn. From 1959, the rate refers to one-year savings deposits, which were introduced in January of that year. Since then no new preferential deposits have been accepted. From July 1, 1963 the same rate has been applied to two-year and three-year deposits.

⁵From International Monetary Fund, International Financial Statistics.

Table 13.

China: Loan Rates of Interest, 1949–69

(In per cent per annum) ¹

Source: Central Bank of China, Taiwan Financial Statistics Monthly.

¹Rates for rediscounts, call loans, and time loans are monthly rates at the end of December converted to annual rate. Free market rate is monthly average rate (based on quotation in Taipei City for the fifth, fifteenth, and twenty-fifth of each month) for unsecured loans converted to annual rate.

²Bank of Taiwan rate through 1960; central bank rate thereafter. The Bank of Taiwan extended credit through call loans prior to the reactivation of the Central Bank. The call loan rate of the Central Bank also applies to its secured advances to other banks.

³Beginning on December 8, 1956, differential rates have been charged by the Bank of Taiwan; these figures give the minimum and maximum rates.

⁴Rate for secured time loans through 1955 and thereafter for secured loans of less than one year.

Table 13.

China: Loan Rates of Interest, 1949–69

(In per cent per annum) ¹

	Central Bank Rate		Time Loans		Average Free Market Rate for Unsecured
Year	Rediscounts	Call loans 2	Bank of Taiwan 3	Other banks 4	Loans
1949	…	46.80	46.80	118.80	208.8
1950	…	41.40	21.60	46.80	144.0
1951	…	41.40	21.60	57.60	126.0
1952	…	36.00	21.60	39.60	79.2
1953	…	21.60	14.40	28.80	51.6
1954	…	21.60	11.88	23.76	49.2
1955	…	21.60	11.88	22.32	54.0
1956	…	18.00	10.80–22.32	21.60	46.8
1957	…	18.00	10.80–22.32	19.80	43.2
1958	…	18.00	10.80–20.32	19.80	39.6
1959	…	18.00	10.80–20.88	18.00	46.8
1960	…	18.00	10.80–20.88	18.00	46.8
1961	14.40	16.20	10.80–18.72	16.20	32.4
1962	12.96	15.84	7.50–18.72	15.84	32.4
1963	11.52	14.04	7.50–16.56	14.04	28.8
1964	11.52	14.04	7.50–15.48	14.04	25.2
1965	11.52	14.04	7.50–15.48	14.04	28.8
1966	11.52	14.04	7.50–14.76	14.04	25.2
1967	10.80	13.32	7.50–14.04	13.32	25.2
1968	11.88	14.04	7.50–14.04	13.32	18.8
1969	10.80	13.32	7.50–14.04	13.32	28.8

Source: Central Bank of China, Taiwan Financial Statistics Monthly.

¹Rates for rediscounts, call loans, and time loans are monthly rates at the end of December converted to annual rate. Free market rate is monthly average rate (based on quotation in Taipei City for the fifth, fifteenth, and twenty-fifth of each month) for unsecured loans converted to annual rate.

²Bank of Taiwan rate through 1960; central bank rate thereafter. The Bank of Taiwan extended credit through call loans prior to the reactivation of the Central Bank. The call loan rate of the Central Bank also applies to its secured advances to other banks.

³Beginning on December 8, 1956, differential rates have been charged by the Bank of Taiwan; these figures give the minimum and maximum rates.

⁴Rate for secured time loans through 1955 and thereafter for secured loans of less than one year.

View Table

Table 13.

China: Loan Rates of Interest, 1949–69

(In per cent per annum) ¹

	Central Bank Rate		Time Loans		Average Free Market Rate for Unsecured
Year	Rediscounts	Call loans 2	Bank of Taiwan 3	Other banks 4	Loans
1949	…	46.80	46.80	118.80	208.8
1950	…	41.40	21.60	46.80	144.0
1951	…	41.40	21.60	57.60	126.0
1952	…	36.00	21.60	39.60	79.2
1953	…	21.60	14.40	28.80	51.6
1954	…	21.60	11.88	23.76	49.2
1955	…	21.60	11.88	22.32	54.0
1956	…	18.00	10.80–22.32	21.60	46.8
1957	…	18.00	10.80–22.32	19.80	43.2
1958	…	18.00	10.80–20.32	19.80	39.6
1959	…	18.00	10.80–20.88	18.00	46.8
1960	…	18.00	10.80–20.88	18.00	46.8
1961	14.40	16.20	10.80–18.72	16.20	32.4
1962	12.96	15.84	7.50–18.72	15.84	32.4
1963	11.52	14.04	7.50–16.56	14.04	28.8
1964	11.52	14.04	7.50–15.48	14.04	25.2
1965	11.52	14.04	7.50–15.48	14.04	28.8
1966	11.52	14.04	7.50–14.76	14.04	25.2
1967	10.80	13.32	7.50–14.04	13.32	25.2
1968	11.88	14.04	7.50–14.04	13.32	18.8
1969	10.80	13.32	7.50–14.04	13.32	28.8

Source: Central Bank of China, Taiwan Financial Statistics Monthly.

¹Rates for rediscounts, call loans, and time loans are monthly rates at the end of December converted to annual rate. Free market rate is monthly average rate (based on quotation in Taipei City for the fifth, fifteenth, and twenty-fifth of each month) for unsecured loans converted to annual rate.

²Bank of Taiwan rate through 1960; central bank rate thereafter. The Bank of Taiwan extended credit through call loans prior to the reactivation of the Central Bank. The call loan rate of the Central Bank also applies to its secured advances to other banks.

³Beginning on December 8, 1956, differential rates have been charged by the Bank of Taiwan; these figures give the minimum and maximum rates.

⁴Rate for secured time loans through 1955 and thereafter for secured loans of less than one year.

On the whole, the experience of an active interest rate policy in China has pertinent lessons for other developing countries, even those that may not have encountered rapid inflation or hyperinflation. First, it shows that there is a critical zone even in a hyperinflation³⁷ within which an appropriate level and structure of nominal interest rates will not only counter the flight from money into goods but also stimulate financial savings. The rationale of the interest rate policy with its primary accent on deposit rates rather than on loan rates derived from the paramount need in a hyperinflation to increase savings by diverting funds from consumption rather than by merely restraining investment. Second, it also emphasizes that an anti-inflationary policy cannot rely merely on the differential between deposit and loan rates unless it also ensures a positive real rate of return to depositors. Consequently, the nominal interest rates were pitched high enough—in fact far above levels that are regarded as conventionally normal—to yield a reasonably attractive real return to the saver. For the greater part of the period 1952-58 the real rates of interest on deposits were positive but well within the conventional range of rates, despite the high nominal rates. Likewise, over the years the gap between the nominal and real rates of interest has been progressively narrowed, attesting to the feasibility of a policy of fixing nominal rates in increasingly closer approximation to the real rates. Third, it highlights the scope of central bank support of an “inverted” interest rate structure consequent upon a high interest rate strategy.

Another feature is the flexibility of interest rate policies in China as evidenced by the subsequent lowering of lending and deposit rates that was associated with a corresponding rise in the ratio of liquid assets, including quasi-money, to GNP. Thus, the ratio increased from 9.6 per cent in 1953 to 37.4 per cent in 1969, whereas the rate of interest on one-year savings deposits declined from 24 per cent to nearly 10 per cent. This suggests that the disincentive effects of falling interest rates were more than offset by the stimulus to the general propensity to save by the rising levels of income and the progressive strengthening of confidence in the currency. Thus, the overall strategy had two broad phases. In the first phase nominal rates were pitched sufficiently high, or at rising levels, to stimulate financial savings. The levels at which financial savings became responsive to interest rate increases may be described as the first stage of “criticality.” This is the point at which the inflationary psychosis may be said to have been effectively countered. But it was obviously not necessary to maintain rates at this level because, once deposits became responsive, the momentum imparted by the initial rise in rates could be depended upon to sustain future increases in deposits, provided that other factors were favorable. The second stage was reached when the impact of falling, or low, rates was more than offset by rising incomes and confidence, so that despite the lowering of rates there was a progressive rise in financial savings.

Although there are no adequate data covering the whole period, there is reason to believe that the share of unorganized finance in China has gradually declined as a result of the policy of realistic rates in the organized sector. A sample survey of the flow of funds ³⁸ of private enterprise in 1968 showed that the share of the unorganized sector (excluding nonfinancial institutions) in the money and capital market had declined from nearly 45 per cent in 1964 to about 38 per cent in 1967. The impact of the high interest rate strategy on the unorganized sector is also indicated by the progressive decline in the average free market rate for unsecured loans over the period 1949–69 from about 208 per cent to 28 per cent.

It is more difficult to evaluate the efficacy of high interest rates in stimulating total private savings in China, even though the ratio of private savings to GNP has improved over the period as a whole from about 3 per cent in 1953 to a range of 10–13 per cent in recent years. There is, however, no consistent trend in the aggregate private savings in relation to GNP. This, of course, follows from the absence of a determinate causal link between interest rate changes in financial media and the aggregate real savings of the private sector. It would not be straining the evidence too much to suggest that a realistic interest rate structure possibly helped to create a more favorable climate for savings in general, i.e., the spillover effects of the markets for financial media were possibly beneficial in terms of stimulating saving as a whole. This relationship is, of course, admittedly difficult to quantify.

Undoubtedly, the success of interest rate policies in China in mobilizing voluntary private saving was helped by the general economic situation, notably, by the availability and effective use of foreign (U. S.) economic assistance³⁹ and the gradual elimination of the foreign exchange gap by the growth of exports. Even so, the recourse to realistic interest rates must be rated as one of the most influential factors in the economic recovery and growth of China. After all “foreign aid, sizeable as it was, would not have produced stability if it had not been accompanied by the self-help measures adopted by the Chinese. Experience shows that countries which pursue reckless fiscal and monetary policies can consume large amounts of foreign aid and still not achieve the degree of stability attained in Taiwan. Among the factors responsible for the success of the economic program on Taiwan was the use of realistic interest rates to encourage voluntary saving, especially in the difficult early years.” ⁴⁰

On the whole, Chinese experience aptly exemplifies the role of high but flexible and realistic interest rate policies as a necessary, even if not a sufficient, condition of effective economic stabilization programs. The risk of a large mass of near-money assets impounded by high rates of interest becoming a pocket of latent inflation has been successfully averted by a combination of price and economic stability as well as rising rates of growth.