(Reprinted from HKCER Letters, Vol.62, November/December 2000)

Savings Bonds as Deposit Insurance

Kam Hon Chu


I. Is Deposit Insurance Necessary?

As the new Millennium unfolds, Hong Kong has identified certain strategic areas for future developments so as to maintain and enhance its status as an international financial centre in a new era. Deposit protection is deemed as one of the strategic mandates.

Against a background that almost all major financial centres have instituted deposit insurance schemes, Hong Kong is under pressure to set up a similar scheme to keep up with the Joneses. However, there is no compelling evidence that deposit insurance is an indispensable element for the success of a financial centre. New York and London had evolved into international financial centres long before deposit insurance was introduced in 1935 and 1982 respectively, not to mention the example most familiar to us - Hong Kong's evolution into a financial centre over the last three decades in the absence of ex ante or explicit deposit insurance.

Nonetheless, we should not be complacent with our achievement and totally deny certain positive functions of deposit insurance. Hayek and Mises were far ahead of time to envisage market competition as an evolution process. Put simply, survival of the fittest means making the right choice at the right time in response to an ever-changing global environment. The questions are: Is deposit protection necessary for Hong Kong to continue its future success in a dynamic and competitive financial world? And if so, what is the least costly means to achieve the goal?

Many economists, regulators and policymakers who support deposit insurance fail to distinguish between a run on an individual bank and a run on the whole banking system. The latter interrupts financial intermediation and the payments system, and it is thus costly to the economy. But a run on an individual bank only does not necessarily have a significant adverse impact on financial intermediation because depositors who run on their bank are likely to redeposit their funds into other well-managed banks. A run on an individual bank plays the role of disciplining the bank for excessive risk-taking (Kaufman 1987). This is, however, not to imply that a run on an individual bank is not costly. These costs include depositors' "shoe leather" costs involved in deposit withdrawals, interruptions to investment and production, damages to the bank-customer relationship which takes a long time to build up, etc. But all these costs in total are less significant when compared with the huge cost of deposit insurance, as revealed by the experience of the Canadian Deposit Insurance Corporation (Carr, Mathewson, and Quigley 1994) and the well-known Federal Deposit Insurance Corporation (FDIC) and Federal Savings and Loan Insurance Corporation (FSLIC) debacles in the United States.

During October and November 1997 when the Asian currency crisis took its toll on Hong Kong, there was a run on merely a small- to medium-sized bank. With its capital and liquidity ratios below the industry's averages, the bank was probably perceived by its depositors as being more of a risk taker than were other banks. The run was apparently not contagious and there was no run on the whole banking system. Available monetary statistics reveal no flight from deposits to currency, as the currency-deposit ratio remained relatively stable. Depositors' maintained confidence in the banking system was also evidenced by the continued growth in deposits, although there was switching of deposits from Hong Kong dollars into foreign currencies, partly reflecting their expectations about the exchange value of the domestic currency during this period of turmoil. The decline in Hong Kong dollar deposits could also be partly attributable to slackened loan demand. In brief, the bank run and the observed currency substitution reflect depositors' rational portfolio choices rather than a banking panic. The claim that depositors are not sophisticated enough to assess banks' performance and hence require protection from deposit insurance is probably an exaggeration, if not a myth.

It is not our place here to resolve the controversial issue of whether or not it is the government's responsibility to protect small depositors. Based on Hochman and Rogers' (1969) concept of redistributive efficiency and on his own concept of Pareto risk-efficiency, Ho (1992) argues that there is a case, based on efficiency grounds, for introducing deposit insurance. On the other hand, Kaufman (1996, p. 30) gives three more pragmatic reasons to justify why deposit insurance for small depositors is desirable: (i) to achieve social externalities because the costs of monitoring banks for small depositors outweigh the benefits, (ii) to reduce systemic risk, as small depositors are the most likely to run on banks, and (iii) to avoid political battling between the government and small depositors when a bank failure does occur. While all these reasons for protecting small depositors are undeniably legitimate, there are more cost-effective alternatives other than deposit insurance by which to achieve the goals. One such alternative available to the Hong Kong Government is to issue small-denomination government bonds similar to Canada Savings Bonds (CSBs).

II. Canada Savings Bonds

As a postwar development of the War Savings Certificates, CSBs were introduced in 1946 as vehicles of savings and investment for Canadians and subsequently became a means for the Canadian Government to finance its deficits and debts. Because they are issued by the Federal Government of Canada, they are perceived as default-free. Unlike other government bonds, they are sold and redeemed at their face value. They can be sold and redeemed through securities dealers, chartered banks, and other financial institutions which, for a fee, act as the government's agents. Therefore, they are essentially demand instruments, although they may have nominal terms of seven to ten years.

These attractive characteristics explain why CSBs are popular, especially among the young and the old. In 1984, the average household in the over-65 age group held 20.1% in deposits and cash and 6.4% in bonds (mainly CSBs). The corresponding figures are 5.7% and 1.4% for the 35-44 age group and 11.5% and 2.3% for the under-25 group.1 These figures are consistent with the Life Cycle Hypothesis of consumption and savings. More important, they reflect the positive role of CSBs as not only short-term liquid assets to protect the old against unexpected changes in income during retirement but also alternative investment vehicles for the young who have more liquidity needs in the early phase of the life cycle.

CSBs are so popular that bank loans to finance the purchase of these bonds at the time of their issue have once accounted for a fairly considerable percentage of personal loans. The chartered banks either make loans to individuals under an official monthly savings plan to buy the new CSBs or to finance the bulk purchases of the bonds by employers who undertake to operate payroll savings plans under which their employees may agree to purchase CSBs on an instalment basis through payroll deductions.

According to a study conducted by McPhail and Caramazza (1990), CSBs and personal deposits are close substitutes as a 1% increase in the constant dollar stock of CSBs reduces M2 by 0.12% and M2+ by 0.09%. There is also evidence indicating that M2+, CSBs and Treasury bills are substitutes (McPhail 1993). Recently, the Bank of Canada has constructed new series of monetary aggregates. The broadest money supply definition is now M2++, which includes Canada savings bonds, currency in public circulation and various types of deposits held at depository institutions. As at the end of May 2000, Canada savings bonds outstanding were C$ 27 billion, or 2.6 % of M2++. All these reflect the importance of Canada savings bonds as a form of "near money."

III. Savings Bonds for Hong Kong

With a history of a high saving rate and fiscal surpluses, it may appear that the Hong Kong Government does not need to issue savings bonds. However, as the economic and fiscal conditions have changed somewhat since the Asian Currency Crisis, savings bonds are an option for the Government as a temporary method of financing its deficits. Furthermore, the fact that such savings bonds can serve to protect small depositors and to prevent contagious bank runs by them should not be overlooked. The idea is closely associated with that of increasing the use of money-market mutual funds and other short-term instruments (Ho 1992, p. 32), but savings bonds, being default-free and capital-risk free, are more appropriate than are these financial instruments for achieving the desired goal. Small depositors can choose to hold their wealth, or part of it, in the form of savings bonds, depending on their degree of risk aversion. When a bank is rumoured or reported to be in financial distress, holders of these savings bonds do not need to participate in a run, because their savings bonds are redeemable at any financial institutions that act as the Hong Kong Government's agents.

It should be reiterated that the savings-bond scheme intends to prevent contagious bank runs by small depositors only and that bank runs by sophisticated depositors could still be possible. However, bank runs should not be viewed as an evil because they are effective market discipline on mismanaged banks, as evidenced by the debacles of Canadian Commercial Bank and Northland Bank in 1985, not to mention the well-known collapse of Continental Illinois National Bank and Trust of Chicago in 1984, among others. These incidents also demonstrate that deposit insurance cannot get rid of bank runs and failures either.

During the last decade, the Hong Kong Monetary Authority (HKMA) has taken a number of measures to develop the debt market in Hong Kong. High-quality Hong Kong dollar debt instruments with a minimum denomination of HK$ 50,000 and original maturity of not less than five years were issued by public companies, such as the Hong Kong Mortgage Corporation (HKMC). The minimum denomination of HK$ 50,000 is about 22% of the median household annual income. In sharp contrast, the minimum denomination of CSB is a meagre 0.1% of the average family income in Canada. Such debt instruments are therefore less liquid and affordable than savings bonds, not to mention their potential capital risk.

How feasible would it be to implement this proposal? One may argue that under such a scheme the Hong Kong Government competes with commercial banks for loanable funds from small depositors. To a certain extent this is correct. But if the Hong Kong Government's main objective in issuing such savings bonds is to protect small depositors and to stabilize the banking system rather than to finance its expenditures or deficits, the loanable funds it absorbs in the form of savings bonds will ultimately flow back into the financial system. The Government in this case acts as a broker or agent, collecting loanable funds from risk-averse small depositors for the banking system. A possible way to minimize the potential moral hazard problem due to the Government's over-involvement is to impose by law a ceiling on the issue of savings bonds. At the same time, the proceeds from issuing savings bonds should also be diversified among banks, not only to minimize risk but also to prevent certain potential problems due to concentration of government deposits in a handful of banks.

Will banks strongly oppose this proposal? If their perennial claim that administering bank accounts with small deposits is unprofitable is true, then their operational costs would be reduced when small depositors switched their deposits to savings bonds. Banks could also supplement their revenues with fees they would receive from the government for administering the savings bonds. Overall, this proposal is expected to have a very minor, if not negligible, impact on bank profitability. More important, it is much less costly than deposit insurance. At a premium rate of 1/6%, deposit insurance in Canada cost member institutions C$ 515 million in fiscal 1998/99, or about 4% of their pre-tax net interest income, a sizeable percentage. The savings-bond scheme is also more equitable because the deposit-insurance premium raises the operational costs of reputable and prudently run banks, although part of the costs can be transferred to depositors. These banks virtually subsidize their less-efficient-and-higher-risk rivals under a flat-rate deposit insurance scheme.

Like other insurance schemes, deposit insurance gives rise to the notorious moral hazard problem. More stringent banking regulation to prevent the moral hazard problem from occurring implies the imposition of a heavier regulatory burden both on banks and on the regulator. The increased regulatory costs will ultimately be shared by depositors and tax payers. In sharp contrast, the savings-bond scheme is much less costly to administer. As the issuer, the government can delegate the responsibility of selling and redeeming savings bonds to banks, a task they can easily handle with their existing capacities with little increases in operational costs.

Will savings bonds be well-received by depositors in Hong Kong? The expected return on the savings bond may not be attractive to most depositors. However, it should be clarified that the objective of the scheme is not to offer a high return, risk-free investment vehicle to depositors. To get a higher expected return, depositors need to take more of a risk. On one hand, the savings-bond scheme provides a market, which may be currently absent in Hong Kong's financial system, for relatively risk-averse depositors to invest in a default-free, highly liquid, and low-denomination financial instrument. On the other hand, it allows market discipline to continue to function and makes possible the avoidance of the moral hazard problem associated with deposit insurance.

IV. Conclusion

As other countries' experiences show, the costs of an explicit, non-risk-based deposit insurance scheme are more likely to outweigh its benefits. Once introduced, the scheme is likely to perpetuate because of political rather than economic reasons. Millions of dollars are spent each year as regulatory costs to mitigate the notorious moral hazard problem, whereas another sizeable amount is invested on studies and proposals to reform the current deposit insurance systems. Before a properly designed deposit insurance scheme is conceivable and to be adopted by Hong Kong, the savings-bond scheme is a low-cost, flexible and reversible alternative as a financial safety net.


1 See The Distribution of Wealth in Canada 1984 by Statistics Canada. A similar survey on household wealth has recently been conducted by Statistics Canada, but the results have not been released yet at the time of writing this paper.



Canadian Deposit Insurance Corporation. Annual Reports, various issue.

Carr, Jack L., Frank Mathewson and Neil C. Quigley (1994) Ensuring Failure: Financial System Stability and Deposit Insurance in Canada. Toronto: C.D. Howe Institute.

Ho, Lok Sang (1992) "Deposit Insurance: An Economic Perspective," Hong Kong Economic Paper 22: 21-33.

Hochman, 1I.M. and J.D. Rogers (1969) "Pareto Optimal Redistribution," American Economic Review 59: 542-57.

Kaufman, G.G. (1987) "The Truth about Bank Runs," pp. 9-40 in C. England and T. Huertas (ed.) The Financial Services Revolution. Washington, D.C.: Cato Institute.

(1996) "Bank Failures, Systemic Risk, and Bank Regulation," Cato Journal 16(l): 17- 45.

McPhail, Kim, (1993) "The Demand for M2+, Canada Savings Bonds and Treasury Bills." Bank of Canada Working Paper 93-8, Ottawa.

McPhail, K., and F. Caramazza (1990) "The Demand for M2 and M2+ in Canada," Bank of Canada Working Paper 90-3. Ottawa.


| Index | Research Projects | HKCER Letters |
| Speaker Program / Conference | Index of Economic Freedom |
The Hong Kong Centre for Economic Research
School of Economics and Finance
The University of Hong Kong
Phone: (852) 2547-8313 Fax: (852) 2548-6319
email: hkcer@econ.hku.hk