(Reprinted from HKCER Letters, Vol. 84 Jul-Dec 2005)
The Political Economy of Interest Rate Deregulation
and Interest Rate Targeting
Kam Hon Chu
The Political Economy of Interest Rate Deregulation and Interest Rate Targeting
Recently, the Hong Kong Monetary Authority (HKMA) announced three refinements to the operation of the Linked Exchange Rate System that aim at promoting stability and smooth functioning. Besides introducing a strong-side convertibility undertaking and a convertibility zone for the Hong Kong dollar, the recent monetary arrangements enable the HKMA to conduct market operations, for example, to remove market anomalies that may arise from time to time. The local financial community has expressed concern that the HKMA's policy stance itself could be a source of uncertainty. As Hong Kong's de facto central bank, the HKMA undoubtedly has the power to move financial markets. When it recently revealed a proposal for a mortgage index to reflect the interest cost in the Hong Kong banking industry, shares of medium and small local commercial banks immediately plummeted in value across the board, resulting in millions of dollars in shareholder losses.
Beginning in 1994, the Hong Kong banking industry gradually phased out the Interest Rate Rules administered by the Hong Kong Association of Banks. The rules had been put into effect in 1964 following the interest rate war that took place in the late 1950s and early 1960s. A primary objective of interest rate deregulation is to promote efficiency through competition, thus strengthening Hong Kong's status as an international financial center. In July 2001, interest rate deregulation was fully implemented, and banks are now free to offer any interest rate they wish to compete for deposits. Although the prime or best lending rate was not under the jurisdiction of the now-defunct Interest Rate Rules, in the past, whenever there were interest rate changes, the two leading banks ¡X Hongkong and Shanghai Banking Corporation (HSBC) and Standard Chartered ¡X would to take the lead in changing their prime rate to remain in tune with deposit-rate changes orchestrated by the Hong Kong Association of Banks. By and large, the interest rate cartel kept interest-rate changes orderly.
The rules of the game have changed since the interest rate liberalization. As expected, interest rates are no longer cartelized but vary from bank to bank. The two leading banks are not always ahead of the pack in changing their interest rates. In the past, those two banks' prime rate was commonly referred to as a proxy for the interest rate level in the local economy. By contrast, the new interest rate structure may appear confusing. For instance, there are at least two prime rates at the time of this writing (July 2, 2005): one at 5.75% p.a. quoted by HSBC, Hang Seng Bank, and the Bank of China, and another at 6% p.a. quoted by other banks such as Standard Chartered Bank and the Bank of East Asia. This situation is further complicated by a banking practice whereby other lending rates are usually based on the prime rate. A well-known example is banks' mortgage rates, which are quoted in terms of hundreds of basis points based on the prime rate; say, P-2, meaning 2% below the prime rate. Perplexing as this structure may seem, however, the current financial environment is by no means disorderly and is certainly not chaotic. After all, in a free-market economy, prices have to change to fulfill their signaling roles in attaining the efficient allocation of resources. As prices for credit or loanable funds, interest rates are no exception. Although the current phenomenon is new to us, it is premature to classify it as an anomaly. What is important is that there are no apparent signs of any failures in the financial markets. On the contrary, there is empirical evidence in support of the Subsidy Reduction Hypothesis, namely that depositors gain from interest rate deregulation because banks' monopoly rents from the cartel are dissipated.1
Nonetheless, the HKMA appears discomfited by the current diversity in interest rates. For years, it has advocated severing the link between the mortgage rate and the prime rate. According to the latest publicly available information, the HKMA is in the process of compiling a mortgage index, called the composite interest rate and abbreviated as C, which is a weighted average of Hong Kong Interbank Offered Rates (HIBOR), fixed deposit rates, and savings deposit rates. The HKMA claims that C is not only a better estimate of the cost of loanable funds than the prime rate (abbreviated as P) but is also less volatile. According to its research results, C changed 12 times during the period January 2001 to April 2005, while P changed 15 times.2 The adoption of C is expected to encourage bankers to pay more attention to the cost of funds when they determine mortgage rates.3
Although indexes are simple and convenient in terms of application and interpretation, most are not sufficient statistics.4 Consequently, there are information losses, and hence undue use of an index comes with a risk of drawing incorrect inferences or making wrong decisions. Therefore, it is unclear why a mortgage index is more desirable than individual banks' interest rates, which, after all, are relative prices that are more relevant to economic decision-making than is the aggregate or average price level.
Besides potential losses of information, a conceptual as well as practical problem in constructing such an index is the difficulty in precisely defining and measuring the input and output of a financial intermediary. Are deposits an input or output of a bank? On the one hand, deposits can be regarded as an input, as banks need a depository base to create loans. But, on the other hand, deposits are an output, because they provide depositors with services such as denomination intermediation services, maturity-risk intermediation services, and other financial intermediation services. In theory, it is well known that banks create secondary deposits whenever they create loans or acquire assets. In practice, however, banks usually require their borrowers, including mortgage borrowers, to hold certain compensating balances in the form of deposits. In sum, the input-output classification is blurred. Therefore, it is not clear whether the mortgage index (namely C) based on various deposit rates and the HIBOR can serve as an estimate of the cost of mortgage lending by banks. Given the fungible characteristic of "money" or loanable funds and the emergence of new financial products, the "true" cost of lending cannot simply be based on the interest cost of the supply of loanable funds. The real resource costs, such as salaries and wages, rent, and so forth, absorbed in the process of financial intermediation have to be taken into account as well.
For the sake of argument, let us suppose that the input-output classification is only a conceptual issue and that the real resource costs are negligible only when compared with interest costs. The question still remains as to whether such an average mortgage index can be appropriately applicable across the board to local banks, which range from international banks that are among the 100 largest in the world to medium and small local banks. Although there are limits to economies of scale and of scope, the mortgage index tends to reflect the conditions of larger banks more than those of smaller banks. This can potentially place the latter at a competitive disadvantage if banks are required to use the index as a basis for pricing mortgages.
Even if most banks are operating efficiently such that no further scale or scope economies can be exploited, the mortgage index cannot be an unbiased estimate of individual banks' operating costs. In the modern finance literature, banks are regarded as information processors to overcome or to alleviate information asymmetry in the financial market. In an economy with heterogeneous economic agents and diverse information, each individual bank's clientele represents a unique set of information such that information cost and risk exposure vary from bank to bank. It follows that interest rates among banks also differ. As the late Nobel Laureate Hayek correctly pointed out 60 years ago,5 economic individuals use practical and personal knowledge to make plans and to reach decisions to identify their best opportunities and ways of living in society. The competitive market is a mechanism that facilitates the process of discovery. As the lackluster performance and ultimate collapse of centrally planned economies has shown, it is a grave mistake to believe that knowledge can be centralized.
It should be clear by now why the law of one price does not hold in banking, unless the financial product is standardized and homogeneous. Like prices of differentiated products, mortgage rates (and loan rates in general) offered by banks do not have to be the same, although they tend not to substantially deviate from each other because of market competition. Differences and changes in prices are not anomalies. Consider the stock market. The Hang Seng Index serves as a timely and convenient indicator to reflect stock market activities. But there is no sound economic reason for why the value of an individual stock should be based on or tied to the index. The same reasoning applies to the proposed mortgage index.
Nevertheless, in the last couple of years the HKMA has advocated abandoning the link between the mortgage rate and the prime rate. But bankers prefer to adhere to their conventional practice. In all likelihood, the proposed mortgage index is another attempt on the part of the HKMA to eradicate the prime rate as a basis for the pricing of mortgage lending. As before, when new measures were introduced, Mr. Joseph Yam, Chief Executive of the HKMA, has stressed that the HKMA will not compel banks to adopt C; banks are free to choose C or P as the basis for setting their own mortgage rates. However, the HKMA's motivation is possibly to bring banks to heel.
But how likely are banks not to adopt C when it is introduced officially? This is a difficult question to answer. Regulation is about not only economic efficiency but also redistribution. In the past, banks as a group gained from the interest rate cartel at the expense of depositors and borrowers. However, not all banks were in favor of the cartel. Some would have liked to abolish it so that they could have more opportunities to expand their market shares through price competition. Now, business conditions have apparently changed, and competition is keen. The desire to benefit from regulation may make some bankers happy to see less competition, if not re-regulation. When there is a supply of regulation, there is also demand for regulation, because some can potentially gain from the regulatory process. At least one local bank has already voiced its support for a standardized basis for setting the mortgage rate, although the majority has so far remained silent on the issue. Even if the HKMA does not enact mandatory adoption of C, it is still possible that the majority of banks will find it profitable to adopt C, because doing so can effectively be a means to form an implicit, unofficial interest rate cartel. If this turns out to be the case, the Hong Kong banking industry will enter a new phase of the regulatory cycle (i.e., from interest rate deregulation to re-regulation). Consequently, some banks, and, more important, the non-bank public will lose the regulatory game. Besides the redistribution of wealth, the efficiency loss from re-regulation may hamper Hong Kong's status as a leading international financial center and may also inhibit the local economy's growth.
The aforementioned shortcomings and potential negative impact of the mortgage index beg the question: what is the objective of introducing such a mortgage index? The resource cost in computing the mortgage index is unlikely to be lower than its counterpart used by the mass media in reporting individual banks' mortgage rates. In terms of information content, the mortgage index, however, cannot be more revealing than individual banks' mortgage rates. In brief, there should be more cost-effective means available other than the mortgage index to closely keep track of changes in funding cost.
Besides the officially stated objectives of reflecting the average cost of funds of banks and helping banks improve interest rate risk management, one possible answer to the question posed above is related to the HKMA's choice of a target and indicator in the formulation and implementation of monetary policy. Nowadays, central banks around the world are using interest rates as one monetary policy instrument to help them achieve their goals. Technically, the HKMA can influence liquidity and hence interest rates in the local interbank market through repo transactions using Exchange Fund bills and notes. But decisions to change the prime rates largely rest on the banks themselves. This is partly because of the huge size of the interbank market relative to the market for Exchange Fund bills and notes and partly because of the openness of the domestic economy.
The Hong Kong government has in general maintained a positive non-interventionism policy stance, although it occasionally effected changes in interest rates with the cooperation of the Hong Kong Association of Banks in the days when an interest rate cartel was in place, notably during the Hong Kong dollar crisis in the early 1980s. Following interest rate deregulation, however, it seems that the practice is no longer a path toward interest rate changes. When, earlier this year, Mr. Yam avowed his concern about the interest rate differentials between the Hong Kong dollar and the greenback and hence about the risk of a full "catch-up" in the local interest rates, local bankers did not toe the line and raise their interest rates accordingly.6 Mr. Yam might well have been embarrassed, even if he was broad-minded enough not to view this as a challenge to his authority.
From the regulator's point of view, moral suasion is, of course, less dependable than an actual policy measure to achieve a goal. Furthermore, the HKMA possibly requires a benchmark interest rate or index, like the Federal Funds Rate in the United States or the Bank Rate in Canada, that can serve as a target to convey to the public information about the Authority's monetary policy stance and at the same time has a substantive impact on credit conditions and other interest rates.7 A mortgage index seems to be an excellent, if not ideal, choice because of its potential impact on one of the largest pillars in the local economy ¡X the property market. Residential mortgage lending accounts for about one-quarter of loans and advances of all authorized institutions (i.e., licensed banks, restricted licensed banks, and deposit-taking companies). An HKMA-engineered change in the mortgage index (and hence banks' mortgage rates) is expected to have a leverage effect on aggregate economic activity. The index's strength, however, can also be its weakness.
The potential weakness arises from rent-seeking activities on the parts of various pressure groups with vested interests in the local property market, and closely related industries. The more transparent and stronger the index's impact on the property market, the higher the level of these activities. Real estate developers, brokers, and the like have incentives to lobby for a lower mortgage rate to induce or to sustain a booming property market, even though banks may find that a higher mortgage rate is necessary for the trade-off between risk and expected return that they face. While the HKMA has maintained a reasonably high degree of transparency and independence, it is uncertain whether it will yield to political pressure after the monetary regime has changed from a relatively passive, rule-based currency-board type system to one in which the HKMA is pursuing an active stabilization policy. This uncertainty or concern does not come out of thin air. First, there is empirical evidence in support of the political business cycle hypothesis in industrial countries, even in countries with a higher degree of central bank independence like the United States.8 Second, economists do not yet fully understand central bankers' motivation and behavior, and it is unclear whether such an understanding is even attainable. While the theory of bureaucratic behavior may appear radical, we cannot entirely rule out the possibility that central bankers pursue their own self-interests rather than those of the public.9
To avoid welfare losses due to rent seeking and to a monetary policy that deviates from desirable social goals, if the HKMA has to implement interest rate targeting or smoothing, it may perhaps consider an index that reflects the level of aggregate economic activity more than that in a particular sector. Such an index can be, for instance, the inflation rate or the Monetary Condition Index, similar to the one pioneered by the Bank of Canada and adopted by several central banks such as the New Zealand Reserve Bank.10 Admittedly, no index is free from distributional bias. For instance, although society as a whole suffers from high inflation, some income groups gain from (unanticipated) inflation, while some lose. Even a broadly based and relatively unbiased target like nominal or real national income may not be acceptable to environmentalists if increased economic growth causes more pollution. But it is hoped that a more broadly based index will have a less-concentrated impact than a less broadly based index on one particular sector and hence will result in fewer incentives for economic agents to form pressure groups to influence the central bank.
Because of the political economy and macroeconomic consequences, the choice of an interest rate target has to be examined carefully before a switch is made to a new monetary regime. The timing of the release of information about the proposed mortgage index is unlikely to be totally unrelated to the most recent changes in the monetary system, such as the introduction of strong-side convertibility undertaking and the proposed transferability between Certificates of Indebtedness and the Aggregate Balance. These all reflect the beginning of a new but expected era in which the HKMA plays a more active role as a de facto central bank, as indicated in Mr. Yam's statement that "we would not attempt to define quantitatively the combination of exchange rate and interest rate conditions which may cause us concern. Nor is it practicable to describe the exact forms of our possible presence, should we not feel convinced that the anomalies are reasonable. Our intentions are, nevertheless, crystal clear: to preserve exchange rate stability implied by the Linked Exchange Rate System and to promote smooth interest rate adjustments implied by that system."11
The HKMA's intention to stabilize the financial markets should be welcomed rather than condemned. But, as Adam Smith correctly pointed out more than 200 years ago in his classic the Wealth of Nations, governments ¡X even those with the best intentions ¡X nearly always serve the public less effectively than does the enterprise of the individual trader. A central bank's means to achieve its goals should not interfere with market activities, particularly when there are no clear signs of any market failures. More important, the history of central banking reveals that a central bank's intended stabilization policy can turn out to be a source of destabilizing the economy, as evidenced by the U.S. Federal Reserve's policy during the Great Depression.12 We all truly wish that Hong Kong would be an exception whenever the HKMA implements its stabilization policy in the new monetary regime.
1 See Kwan, S. (2003) "Impact of Interest Rate Deregulation in Hong Kong on the Market Value of Commercial Banks," Journal of Banking and Finance 27: 2231-48.
2 The findings that C more accurately reflects the cost of loanable funds than P on the one hand and that the former is also less volatile than the latter on the other are somewhat counterintuitive, if not inconsistent, at least in this author's mind. As HIBOR changes frequently in response to changing market conditions, C should be more volatile than P. In fact, the alleged lower volatility of C is due to an ad hoc criterion whereby C is adjusted only when it deviates from the actual figure by more than 0.25%. Apparently, one can further decrease the volatility by arbitrarily choosing a larger deviation as the adjustment criterion. More important, it is unclear to me how the adjustment to C will occur. Will the HKMA step into the financial market to smooth out the interest rate whenever the deviation is larger than 0.25%, such that the announced or published C and the actual rate are equalized? If so, this implies that the HKMA will play an active role in interest rate smoothing. Or will the published C automatically adjust to the actual rate when the criterion is fulfilled? If this is the case, commercial banks will lose some degree of autonomy and timeliness in setting their own mortgage rates when they use C as a basis. This is because they will have to rely on the HKMA for computing and publishing C. In either case, the HKMA's involvement will be, in one way or another, greater than it is currently.
3 Based on the information revealed by Mr. Peter Pang, Deputy Chief Executive of the HKMA, in an interview with a leading local newspaper Ming Pao reported in the June 22, 2005 issue.
4 A statistic is said to be sufficient if it loses no information about the unknown parameter contained in the sample. For more technical details, see for example Mood A., F. Graybill, and D. Boes (1974) Introduction to the Theory of Statistics, 3rd edition, McGraw-Hill.
5 Hayek, F.A. (1945) "The Use of Knowledge in Society," American Economics Review 35: 519-30.
6 As a central banker, Mr. Yam's concern about the interest rate risk arising from the interest rate differential is totally legitimate, as interest rates tend to move together in the same direction, especially against the backdrop of financial globalization and liberalization. It is correct that interest rates in Hong Kong are expected to catch up with U.S. interest rates sooner or later because of prevailing interest rate differentials. However, the argument for a full catch-up is implicitly based on the empirical validity of the covered interest rate parity (CIP). There is considerable empirical evidence supporting the CIP for financial markets where short-term assets are comparable in all aspects except currency of denomination, where trade volume is high, and where information and transaction costs are low. The empirical evidence does not bode well for transactions covering long terms to maturity, because financial institutions generally do not have an elastic supply of arbitrage funds for long periods. A visual examination of the time-series movements of lending rates (or other interest rates such as time deposit rates or even money market rates) in Hong Kong and the United States over the last two decades or so suggests that catch-ups were commonly observed but that full catch-ups were more of an exception than the rule.
7 From time to time, The HKMA determines the "base rate," the interest rate on accessing the HKMA's discount window for Hong Kong dollar liquidity. The base rate can fulfill quite satisfactorily the first criterion as a signal but probably less so the second criterion, partly because of the passivity and the volume of discounting activities.
8 For example, the significant influence of the timing of a U.S. election on money growth is documented by Grier, K (1989) "On the Existence of a Political Monetary Cycle," American Journal of Political Science 33: 376-89.
9 Mr. Yam's public service record has been impeccable and his integrity is not in doubt, but there is no guarantee that an elected politician or appointed central banker will not put his or her self-interest ahead of the public interest.
10 The Monetary Condition Index (MCI) is a weighted average of the short-term interest rate and the trade-weighted exchange rate. The Bank of Canada does not set a target path for the MCI but uses it as an operational guide for the impacts of the two underlying economic variables on aggregate demand and price stability. For details, see Freeman, C. (1995) "The Role of Monetary Conditions and the Monetary Conditions Index in the Conduct of Policy," Bank of Canada Review Autumn 1995: 53-60.
11 Mr. Joseph Yam's viewpoint article "Refinements to the Linked Exchange Rate System," dated June 16, 2005.
12 For details, see Friedman, M. and A.J. Schwartz (1963) A Monetary History of the United States, 1867-1960, Princeton, N.J.: Princeton University Press.
Kam Hon Chu is Associate Professor at Department of Economics, Memorial University of Newfoundland.