Information about Asia and the Pacific Asia y el Pacífico

Chapter 8. Can the Riel Prosper? Welding Financial Development and Dedollarization

Olaf Unteroberdoerster
Published Date:
February 2014
  • ShareShare
Information about Asia and the Pacific Asia y el Pacífico
Show Summary Details
Phurichai Rungcharoenkitkul

Cambodia has one of the most dollarized economies in the world. While dollarization has played some part in anchoring the country’s macroeconomic stability and sustaining economic growth over the last two decades, whether it provides the right framework is highly contentious.

This chapter discusses how Cambodia’s dedollarization objective is interlaced with the goals of developing its financial sector and, ultimately, pursuing greater monetary policy effectiveness. It argues that a carefully designed and sequenced introduction of financial and market development initiatives, accompanied by regulatory measures and clear communication about the policy path, could minimize any unintended setback on dedollarization and enable the country to achieve its dual objective.

Apart from the fact that dollarization was not Cambodia’s official choice to begin with, the costs to the country associated with it are high, as shown in Chapter 3. First, the loss of seigniorage revenue alone is likely to be non-negligible.1 Second, the lack of monetary independence limits the authorities’ ability to respond to shocks, a particularly costly constraint considering the economy’s rapid integration with the region and the world and the increasingly volatile global capital movements. Third, the fact that banking sector balance sheets are largely denominated in dollars also limits the authorities’ role as a lender of last resort in addressing financial stability risks. Moreover, Cambodia has come a long way in rebuilding an economic and financial infrastructure capable of maintaining macroeconomic stability in the context of independent monetary policy.

As a result, a longstanding ambition, shared by both the government and the central bank, has been to dedollarize the economy.2 At the same time, with the economy growing rapidly, the demand for efficient financial services and further financial development is also building. Moreover, monetary policy cannot be effective without a developed financial market that helps transmit the policy actions to economywide monetary conditions. With the balance sheets of Cambodian banks being almost entirely in U.S. dollars, making further progress in financial development is easiest done by promoting dollar-based financial activity, a seeming conflict with the dedollarization objective.

Against this background, we start from the basic premise that the critical ingredients for a successful transition toward an independent and market-based monetary policy framework consist of both making significant progress in dedollarization and developing financial markets to create an effective monetary transmission mechanism. In order to reconcile the dual objectives of dedollarization and financial development, the chapter discusses strategic options and their relative merits and highlights the importance of a clear communication strategy to support them.

Toward Greater Monetary Policy Effectiveness

Dollarization in Cambodia occurs both in terms of currency substitution and asset substitution. Under the former, the U.S. dollar widely serves the dual functions as a medium of exchange and as a unit of account in everyday transactions by households and businesses.3 At the same time, with foreign currency deposits accounting for more than 90 percent of total deposits, there is also an extensive dollarization in the interest-bearing assets. This asset substitution on the part of households leaves banks with dollar-dominated liabilities, which are perpetuated when they extend loans mostly in dollars to hedge against currency risks. Both sides of banks’ balance sheets are therefore denominated largely in dollars, which have become an accepted medium of exchange among banks. The literature sometimes refers to the phenomenon as financial dollarization. The current structure of dollarization in Cambodia is summarized schematically in Figure 8.1.

Figure 8.1Different Guises of Dollarization

Different aspects of dollarization help reinforce each other. Widespread currency substitution at the level of firms and households makes it easier for asset substitution and financial dollarization to flourish, and vice versa. Because of this multifaceted and interdependent nature of Cambodia’s dollarization, the strategy for pursuing monetary independence cannot be approached separately through financial market development. For example, while developing an interbank market in U.S. dollars would help promote financial development, it carries a risk of further entrenching financial dollarization and consequently currency substitution. Whether such an initiative would constitute a step forward in the quest for an independent monetary policy framework is therefore unclear.

Meanwhile, financial development is both an integral part of broader economic development and a necessary foundation for effective monetary policy transmission. Without financial development, an independent monetary policy and financial intermediation in general would be ineffective. Short of a significant reduction in dollarization, financial development would do little to enhance monetary independence (and possibly quite the opposite). Therefore, to achieve the ultimate goal of effective and independent monetary policy, Cambodia needs to make progress in both dedollarization and financial development.

A body of cross-country evidence has highlighted the roles of macroeconomic stability, good governance, and institutional quality in helping economies dedollarize (see De Nicolo, Honohan, and Ize, 2003; Yeyati, 2005; and Asel, 2010; for example), but few studies have looked at the relationship between the two objectives of dedollarization and financial development. Country-specific anecdotes, however, confirm potential linkages between the two. For example, the development of a local currency government bond market has been identified as a contributing factor to the successful dedollarization in Peru (Garcia-Escribano, 2010).

While macroeconomic stability is a necessary condition for successful dedollarization, it is not sufficient. With the second objective of financial development, it becomes even clearer that macroeconomic stability must be augmented by other initiatives that are tailored to Cambodia’s circumstances. We propose in this chapter that the strategy to weld dedollarization to financial development be guided by the following general considerations:

  • The strategy should spell out in advance the sequencing of initiatives that together make up a credible plan for dedollarization as well as a commitment to making steady progress in laying the foundation for an independent monetary policy framework.

  • The dedollarization strategy should be credible and comprehensive and aim to address all aspects, including currency substitution, asset substitution, and financial dedollarization.

  • Financial development initiatives should proceed but, wherever possible, avoid perpetuating the degree of asset substitution and financial dollarization. When financial development and dedollarization come into conflict, measures should be introduced to lessen the tension between the two objectives.

Financial Development Initiatives

What are the potential frictions between dedollarization and financial development? In Cambodia’s case, the creation of an interbank market and of a stock exchange with its potential spillovers to the nascent foreign exchange market are the most relevant examples of how financial development initiatives impact progress toward dedollarization. This section reviews both development initiatives in detail and discusses their relationship to dedollarization.

Interbank Market

A new initiative is the development of a collateralized interbank market. Currently, qualified commercial banks have access to a fixed-rate deposit facility at the National Bank of Cambodia (NBC). While the facility is available in both riels and dollars, the extent of financial dollarization means that these deposits are almost entirely denominated in U.S. dollars. The initiative aims to collateralize them by introducing negotiable certificates of deposit (NCDs), which can be traded and used as collateral for interbank lending through repurchase agreements. It is hoped that collateralization will help alleviate counterparty risks, reduce banks’ precautionary motives for maintaining excessive liquidity, and promote an efficient interbank market and commercial banking activities. Ultimately, a fully developed interbank market could serve as the platform for implementing open market operations.

Specific features of NCDs constitute a critical step of the strategic plan. As discussed above, the choice of currency denomination, for both pricing and settlement, is particularly relevant for the tension between financial market development and dedollarization goals. A closely related issue is whether the NBC should securitize the deposits outright or issue entirely new securities and gradually phase out the existing fixed deposit facility. Under the first approach, the NBC would effectively have to denominate the securities in U.S. dollars by default, since nearly all current deposits are in U.S. dollars. Under the second approach, the NBC could issue new securities in riels as well as in U.S. dollars, but the demand for riel-denominated securities is likely to be very low in the near term given Cambodia’s high degree of dollarization.

Regardless of currency denomination, the NBC must also ensure that the new securities will be preferred by banks to the currently available fixed deposits, for example by announcing a timetable for phasing out the fixed deposits. Although the NCDs may command a collateral premium or pay higher coupons relative to fixed deposit rates, the market prices of these tradable securities could adjust to reflect their inherent values. In any case, the two distinct liquidity management instruments remain imperfect substitutes owing to their different short-term volatilities. Fixed deposits provide deterministic cash flows, paying a fixed rate of interest at maturity but incurring a penalty for premature withdrawals, which could give rise to a “kinked” payoff. On the other hand, NCDs would likely provide a smoother payoff profile than fixed deposits, but are subject to market risks.

These nuances mean that (i) the two instruments may attract demand from different types of holders, creating a form of market segmentation, and (ii) the relative demand may depend on prevailing market conditions. For banks with more frequent liquidity needs and higher market-risk tolerance, the NCDs offer greater benefits compared to fixed deposits owing to their collateral value and smoother payoffs. For others, the market risks associated with the NCDs are more binding, making them less attractive. Similarly, during periods of high market volatility and/or low trading liquidity, the fixed deposit facility may become more attractive.

Capital and Foreign Exchange Markets

The effect of market design on financial dollarization is also evident in the case of the Cambodia Securities Exchange, which began trading its first and only stock in April 2012. The stock market was designed to allow transactions to be settled in both dollars and riels, although the listing is required to be in the local currency only. Available information suggests that all trading activity of the debut issue (stock of the Phnom Penh Water Supply Authority) so far has been settled exclusively in riels. A combination of factors may have contributed to this outcome. First, given the riel quotation, any settlement in a foreign currency necessitates a second agreement in the exchange rate, adding to the complexity and risk of the trade. Second, if no exchange rate is agreed among the counterparties within two days of the transaction, the settlement is in riels, as dictated by regulation. That the listing requirement in local currency can effectively encourage greater use of the riel in settlement should serve as an important lesson.

Developing a wholesale foreign exchange market will also become more pressing as the economy makes further progress in financial dedollarization and in developing financial markets, given the attendant growth in and demand for, and the supply of, riels. As recommended by the IMF during the 2012 Article IV consultation, the NBC should scale down its direct involvement in the foreign exchange market, organize a wholesale market with large commercial banks at the core, and allow the market to clear itself. With that market infrastructure in place, wholesale foreign exchange activity should grow in line with the gradual progress in financial dedollarization. When the market becomes sufficiently deep and liquid, the next steps of financial development, including the introduction of a local currency bond market, could take place without disrupting the availability of foreign exchange.

In addition, a foreign exchange hedging tool could provide supplemental incentives for banks to channel excess U.S. dollar liquidity into riel-denominated NCDs. By engaging in sell-buy dollar swap contracts, banks could guarantee themselves cash flows in dollar terms. The riel-denominated NCDs combined with swap contracts could effectively serve as convertible securities, which might be in high demand even by banks with excess dollar liquidity.

However, with virtually no activity for wholesale transactions, it is unlikely that the private swaps market would automatically emerge. Indeed, the limited riel liquidity may not be sufficient to support the buy-sell leg of the swap contract. The NBC could consider bridging this gap by offering an open swaps facility, available only to buyers of riel-denominated NCDs, and by acting as the primary lender of riels to promote the new securities. Against the potential benefits of such an initiative, however, one must also take into account the added market complexity, pricing difficulties, uncertain fiscal burden, and credibility costs.4

Sequencing The Initiatives

The choice of policy initiatives at the disposal of the NBC to drive financial development largely depends on the pace and depth that the authorities choose to undertake dedollarization. The initiatives, as shown in Table 8.1, can be divided into three components: (i) financial development policies, (ii) monetary policy tools, and (iii) regulatory and other measures. These initiatives are collectively the nuts and bolts that, if introduced and implemented in the right sequence, will combine to create an efficient financial market infrastructure and an independent monetary policy framework. Several alternative sequences are outlined next to highlight the issues involved in rolling out these initiatives. This is followed by a discussion of communications strategy, something that is essential to overcome potential coordination failure and to induce households and firms to jointly make greater use of the local currency.

Table 8.1Types of Initiatives
Financial development
  • Stock market

  • Collateralized interbank market

  • Wholesale foreign exchange (FX) market

  • Local currency bond market

  • Basic derivatives

Monetary policy
  • Reserve requirement (RR)

  • Exchange rate policy

  • Liquidity management/open market operations (OMO)

  • Communication by monetary policy committee (MPC)

Regulations/ Other
  • Pricing/settlement convention

  • Remuneration on required reserves

  • Remuneration on fixed deposit facility

  • Deposit insurance

Strategy 1: A Dual Approach

Under Strategy 1, the first milestone would be the introduction of NCDs as new securities in both currencies, with a phase-out schedule for the existing fixed deposit facility (see Figure 8.2). The announced commitment to phase out the old facility should help develop the new interbank market by promoting subscription and liquidity in the secondary market, especially at the early stage. Market activity for NCDs in dollars is expected to dominate initially, and this may help sustain or even increase the already extensive degree of financial dollarization. The government should therefore state at the outset its comprehensive dedollarization strategy and clarify that, once the economy makes sufficient progress in dedollarization, the dollar-denominated NCDs will no longer be issued.

Figure 8.2Strategy 1: Dual approach

Note: CR = Cambodian riel; FX = foreign exchange; MPC = monetary policy committee; NCDs = negotiable certificates of deposit; OMO = open market operations; RR = reserve requirement.

The second milestone of this strategy would comprise a number of specific incentives to promote and strengthen the credibility of the overall dedollarization plan. The key objective at this stage is to propel the economy away from the current state of near-complete dollarization and create a significant boost of riel volume in the financial system. As a crude benchmark, the target could be to achieve riel deposits of 20 percent of total deposits, a level similar to that in other highly dollarized economies.

Key measures would include (i) widening the reserve requirement’s ratio gap between foreign and domestic currencies;5 (ii) remunerating required reserves for riel deposits and lowering remuneration for dollar deposits;6 and (iii) providing deposit insurance to riel deposits. The first two measures are taxes on dollar intermediation and could serve to reduce financial dollarization. The third initiative strengthens the riel’s function as a store of value, helps reverse asset substitution, and also ultimately promotes currency substitution. The potential cost of the latter measure should initially be low, in line with the amount of riel deposits in the system, whereas any incentive distortion (e.g., with regard to competition among banks) could be curbed by limiting the amount of deposit insurance provided.

A third milestone would be to put in place a wholesale foreign exchange market infrastructure, capable of matching sizable flows among institutional participants. As the degree of financial dollarization is progressively reduced, the volume of larger-scale foreign exchange transactions will rise. As the volume for NCDs in riels rises, there will be an opportunity to develop active foreign exchange forward and swap markets. The availability of foreign exchange hedging instruments, by lowering exchange rate risk, should further raise demand for riel-denominated securities. The government could possibly accelerate the pace of financial dedollarization by gradually phasing out the dollar-denominated NCDs, allowing riel-denominated NCDs to serve as the collateral for interbank lending in dollars, with appropriate haircuts or other hedges against embedded exchange rate risks.

As both asset and cash substitution begin to reverse, the NBC can accommodate higher structural demand for riels by calibrating foreign exchange policy to curb excessive appreciation of the currency. This is an opportunity for the NBC to accumulate more foreign reserves without adding to inflationary pressure, which can be achieved by ensuring that the injection of riel liquidity is gradual, consistent with market demand, and does not lead to large swings in the money market interest rate. A small appreciation of the riel could also be allowed, which would lend further impetus to dedollarization, creating a self-reinforcing process. At this stage, the NBC would also need to begin managing riel liquidity actively through a repurchase operation involving riel-denominated NCDs in order to tighten control over short-term interest rates, thus laying the groundwork for possible open market operations.

Once the riel money market and the foreign exchange market are both well established, the next step is to introduce longer-maturity government securities. Developing the public debt market in domestic currency and extending the yield curve would help promote the development of other financial markets (including basic derivatives such as interest rate swaps), which have proven to be an important catalyst for dedollarization in other countries such as Peru (see Garcia-Escribano, 2010). A yield curve would also serve as a transmission mechanism for market-based monetary policy. To this end, the NBC should prepare for such a transition by formalizing a monetary policy framework (objectives, instruments, and targets). By explicitly tying monetary policy to a domestic nominal anchor, the new framework will lend further credibility to macroeconomic policy management, another boost to dedollarization. After sufficient progress is made, the settlement of all banking transactions in dollars could then be terminated.

The key risk of Strategy 1 is that the measures introduced in the first step could fail to induce a substantial shift away from the near-complete dollarization. Falling short of making some progress in this crucial step, the dollar-denominated NCDs could end up bolstering financial dollarization. In that case, phasing them out subsequently might not only be a step backward for financial development, but also constitute a setback to the authorities’ credibility. Issuing NCDs in dollars could also cast doubt on their commitment to dedollarize the economy.

Strategy 2: Dedollarization as Prerequisite

Strategy 2 considers an alternative sequencing, in which only the riel-denominated NCDs are issued initially as new securities. At the start, the demand for the riel-denominated NCDs is likely to be very low, mirroring banks’ current portfolio allocation. Consequently, banks may continue to manage their dollar liquidity through the existing fixed deposit facility. Because banks do not have alternative means of managing their dollar liquidity, the NBC may have no option but to keep the old facility available, until the volume for NCDs in riels has risen sufficiently to be useful as collateral for interbank activity in dollars. However, without an explicit phase-out period, banks would have little incentive to stop relying on the fixed deposit facility and switch to subscribing to NCDs in riels. Thus, money market development might be a more protracted process under this strategy, requiring a more forceful implementation of accompanying dedollarization measures (second milestone under Strategy 1) to further the process of financial development (see Figure 8.3).

Figure 8.3Strategy 2: Riel NCDs only

Note: CR = Cambodian riel; FX = foreign exchange; MPC = monetary policy committee; NCDs = negotiable certificates of deposit; OMO = open market operations; RR = reserve requirement.

What the second strategy buys is insurance that the financial development plan does nothing to jeopardize the dedollarization policy, since the financial markets are developed exclusively in riel denomination. The flip-side is that successful dedollarization becomes a precondition for further progress in financial development. Under the second strategy, only when the extent of financial dedollarization sufficiently declines will the volume of NCDs rise, enabling the riel-denominated NCDs to serve as collateral for both riel- and any remaining dollar-denominated interbank loans. With an interbank market reasonably active in both currencies, a wholesale foreign exchange market could then be developed along with foreign exchange hedging tools. The ability to hedge foreign exchange risks would raise the potential of using riel-denominated NCDs as collateral for dollar-denominated loans and the fixed deposit facility could then be phased out gradually. With an interbank market firmly established, the next steps of developing a debt market and creating a new monetary policy framework can then proceed as in Strategy 1. As discussed earlier (under capital and foreign exchange markets), the NBC could augment the initial stage of this approach by introducing an official swap facility that hedges foreign exchange rate risk for buyers of NCDs in riels. The design of such a facility itself poses important challenges (e.g., regarding term and rate structure), but the facility could be a catalyst for accelerating the demand for riel-denominated NCDs.

A Gradual Approach

The two strategies discussed so far represent distinct cases that may not be optimally aligned with the NBC’s preferences regarding the two objectives of dedollarization and financial development. For example, in practice the authorities may want to ensure the success of dedollarization with a very high probability, but at the same time they may be unwilling to trade off the postponement of financial development indefinitely.

A third strategy could combine Strategies 1 and 2. Under this hybrid strategy, the NCDs could initially be issued exclusively in riels, similar to Strategy 2. However, if no material progress were subsequently observed in interbank market development, a small amount of dollar-denominated NCDs could be issued to boost market activity. The issuance size of NCDs in dollars could be capped below the full market demand and be allowed to increase only gradually and only to the extent that it did not exacerbate the degree of financial dollarization. The fixed deposit facility could continue to exist, but a limit would need to be placed on its size, too, which in turn could be reduced over time.

Under this hybrid strategy, the authorities would retain firm control over the dedollarization objective. The strategy would send a strong signal about their commitment, afford a more gradual transition from the old to the new regime, and still allow greater benefits from financial development. During the gradual transition, any effect on dollarization could also be monitored. Should the degree of dollarization intensify or remain high, the authorities would need to consider strengthening the incentive measures to encourage greater use of the riel and/or scale down the issuance of NCDs in dollars. The opportunity to fine-tune policy measures as warranted without compromising policy credibility is another benefit of this approach.

The key challenge in this case is still to break away from the near-complete dollarized state, so the implementation of broader dedollarization policies through regulatory and other policy tools should remain a top priority. Without these measures, there would be little impetus for the demand of riel securities to take off, and a large-scale financial dollarization would likely persist.


The dollarization phenomenon and the general choice of currency are partly driven by coordination failure. While a systemwide switch to the Cambodian riel may be beneficial for the society as a whole, each individual has little incentive to switch unless others are also willing to change their choice of currency. The perceived likelihood of a successful dedollarization is not only a measure of policy credibility, but could potentially be self-fulfilling.

The authorities can leverage the self-fulfilling nature of public expectations and lend further support to the dedollarization process by communicating clearly and strategically about their plans and objectives. Some key messages to be conveyed include the following:

  • The NBC is committed to dedollarization in a deep and stable financial system based on the riel and to an independent monetary policy that can propel the economy into its next stage of development.

  • The process of financial development is itself driven by a set of carefully sequenced and comprehensive initiatives to achieve this objective.

  • The initiatives will encourage greater use of riels both for ordinary transactions and within the financial system, broadening the existing policy scope, which has so far been focused on dedollarizing the real sector.

  • Both individual households and financial institutions can benefit from greater asset diversification toward riel use, given the authorities’ commitment (i) to keep the local currency strong through stable macroeconomic conditions; and (ii) to develop the financial markets with the riel playing a prominent role. Additionally, risk-adjusted riel funding costs would likely fall.


Financial market development and dedollarization are both important policy objectives in Cambodia that cannot be pursued independently of each other. Rather than prioritizing one objective over the other, this chapter discusses how an appropriate sequencing of measures within a coherent strategy can help reduce the tension and pursue both financial development and dedollarization simultaneously. A fully developed strategy would serve as a concrete policy vision for the NBC in moving toward an efficient and independent monetary framework, while at the same time helping to enhance policy credibility.


    AselIsakova2010Financial Sector Development and Dollarization in the Economies of Central Asia,Unpublished paper (London: European Bank for Reconstruction and Development; Center for Economic Research and Graduate Education—Economics Institute).

    De NicoloGianniPatrickHonohan and AlainIze2003Dollarization of the Banking System: Good or Bad?Policy Research Working Paper No. 3116 (Washington: World Bank).

    De ZamaróczyMario and SopanhaSa2002Macroeconomic Adjustment in a Highly Dollarized Economy: The Case of Cambodia,IMF Working Paper No. 02/92 (Washington: International Monetary Fund).

    Garcia-EscribanoMercedes2010Peru: Drivers of Dedollarization,IMF Working Paper No. 10/169 (Washington: International Monetary Fund).

    IzeAlain and Eduardo LevyYeyati2003Financial Dollarization,Journal of International Economics Vol. 59 pp. 32347.

    Royal Government of Cambodia2012Financial Sector Development Strategy 2011–2020” (Phnom Penh).

    Yeyati E.Levy2005Financial Dollarization: Evaluating the Consequences,Presentation to the 41st Panel Meeting of Economic PolicyLuxembourgApril15162005.

Cited as “a main cost of high dollarization in Cambodia,” de Zamaróczy and Sa (2002) calculated that, as of 2001, lost seigniorage revenue in Cambodia ranged between 0.7 percent and 1.5 percent of GDP.

The government has stated that “Promotion of the riel has been adopted as government policy over the short to medium term, and dedollarization as the long-term goal” (Royal Government of Cambodia (2012, p.9).

Using a portfolio balance approach, Ize and Yeyati (2003) show that the degree of currency substitution is an increasing function of inflation volatility, but a decreasing function of exchange rate volatility. However, because in Cambodia’s case the dollar also serves as a unit of account and households’ local consumption basket is to a significant degree denominated in dollars, the basic portfolio view need not hold. Instead, in this context, the degree of dollarization should increase with the expected depreciation of the local currency, as the latter would lower the purchasing power of the riel-denominated portfolio.

Since there is no benchmark for market swap rates, the NBC would need to dictate the terms of the swaps. A more favorable rate could sweeten the NCDs and boost demand for them, but at a fiscal cost in terms of the impact on the NBC’s balance sheet. By taking the buy-sell leg, the NBC could also stand to gain from any depreciation in riels, which might adversely affect the NBC’s credibility in promoting the stability of the local currency.

The prevailing economic developments would dictate how the widening should be engineered. If there is a need for overall monetary policy tightening, the NBC could raise the reserve requirement relatively more for foreign currency deposits. On the other hand, a more policy-neutral option would be to simultaneously raise the reserve requirement for foreign currency and lower that for riel. Currently, there is ample room for lowering the riel reserve requirement, which stands at 8 percent.

As of June 2013, the reserve requirement on foreign currency deposits was 12.5 percent. Any required reserve for dollars in excess of the pre-2008 level of 8 percent is remunerated at half the SIBOR rate, with the rest not remunerated. The remuneration on required reserves is a relatively recent initiative, adopted when the reserve requirement on foreign currency deposits was raised for the first time in 2008.

    Other Resources Citing This Publication