Abstract
This paper provides an overview of indicators that can be used to illustrate and analyze liquidity developments in financial markets. The measures include bid-ask spreads, turnover ratios, and price impact measures. They gauge different aspects of market liquidity, namely tightness (costs), immediacy, depth, breadth, and resiliency. These measures are applied in selected foreign exchange, money, and capital markets to illustrate their operational usefulness. A number of measures must be considered because there is no single theoretically correct and universally accepted measure to determine a market's degree of liquidity and because market-specific factors and peculiarities must be considered.
I. Introduction
Liquid markets are generally perceived as desirable because of the multiple benefits they offer, including improved allocation and information efficiency. They (i) allow a central bank to use indirect monetary instruments and generally contribute to a more stable monetary transmission mechanism; (ii) permit financial institutions to accept larger asset-liability mismatches, both regarding maturity and currency, thus fostering more efficient crisis management by individual institutions, and reducing the risk of the central bank having to act as lender of last resort for solvent but illiquid credit institutions; and (iii) render financial assets more attractive to investors, who can transact in them more easily.2 The latter benefit, however, may not be true for investors collectively. As Keynes noted (1936, p. 160): “For the fact that each individual investor flatters himself that his commitment is “liquid” (though this cannot be true for all investors collectively) calms his nerves and makes him much more willing to run a risk.” Therefore, recent crises in financial markets, in particular, have triggered studies on how to better judge the state of market liquidity and ideally to better predict and prevent systemic liquidity crises.3
This paper has two main purposes. First, it provides an overview of the many different concepts related to liquid financial markets. Second, it identifies some simple quantitative indicators that Financial Sector Assessment Program (FSAP) missions of the International Monetary Fund (IMF) and the World Bank can use to illustrate the changing degree of liquidity in financial markets over time.4 Finally, the paper also briefly discusses the possibility of creating a composite measure of the “liquidity stance” in financial markets. This latter task was prompted by analysts, such as Borio (2000), who reports that in the run-up to financial crises, markets often appear artificially liquid, but during periods of stress, liquidity tends to evaporate.
The microeconomic concept of liquidity is multifaceted. Market participants perceive a financial asset as liquid, if they quickly can sell large amounts of the asset without adversely affecting its price. Liquid financial assets are thus characterized by having small transaction costs; easy trading and timely settlement; and large trades having only limited impact on the market price. Most of the existing literature gauging liquidity has focused on different dimensions of liquidity of individual financial assets.5 It is generally concluded (Baker, 1996, p. 1) that there: “… is no single unambiguous, theoretically correct or universally accepted definition of liquidity.” Moreover, the importance of some of the characteristics of liquid markets may change over time. For instance, during periods of stability, the perception of an asset’s liquidity may primarily reflect transaction costs. During periods of stress and significantly changing fundamentals, prompt price discovery and adjustment to a new equilibrium becomes much more important.
Liquid markets tend to exhibit five characteristics: (i) tightness; (ii) immediacy; (iii) depth; (iv) breadth; and (v) resiliency. Tightness refers to low transaction costs, such as the difference between buy and sell prices, like the bid-ask spreads in quote-driven markets, as well as implicit costs. Immediacy represents the speed with which orders can be executed and, in this context also, settled, and thus reflects, among other things, the efficiency of the trading, clearing, and settlement systems. Depth refers to the existence of abundant orders, either actual or easily uncovered of potential buyers and sellers, both above and below the price at which a security now trades. Breadth means that orders are both numerous and large in volume with minimal impact on prices. Box 1 illustrates the difference between depth and breadth. Resiliency is a characteristic of markets in which new orders flow quickly to correct order imbalances, which tend to move prices away from what is warranted by fundamentals. These terms reflect different dimensions of the extent to which an asset quickly and without significant costs can be transformed into legal tender.
These terms, however, are also to some extent overlapping. Most of the available data do not fully correspond to these dimensions, which complicates their measurement. In addition, a number of qualitative factors must be considered, since they affect the above-mentioned dimensions of liquidity. They range from the microstructure of the market, the central bank’s implementation of its monetary policy, to risks in the payment and securities clearance and settlement systems.
Indeed, understanding the microstructure of the market is important, when proxies, like bid-ask spreads and turnover ratios, are used as liquidity indicators. A market can be a physical location, an electronic or other platform that allows potential buyers and sellers to interact. Most academics have a neoclassical phantom world in mind with a Walrasian auctioneer performing a price tâtonnement process ensuring instantaneous trading at market clearing prices. In short, prices are a sufficient statistic. In the practitioner’s world, however, trading may take place in various platforms (for instance, dealer or auction markets) at nonmarket clearing prices because of factors such as market illiquidity.
In a dealer market, trading is quote driven. Dealers quote bid and ask prices and may take positions. Therefore, it is often argued that dealers provide liquidity, since they provide a
Illustration of Market Depth and Breadth
Size of existing or easily uncovered bids
Size of existing or easily uncovered bids
| Market: | 1 | 2 | 3 | 4 |
|---|---|---|---|---|
| Bid price | Thin and Shallow | Thin but Deep | Broad but Shallow | Broad and Deep |
| $50 | 100 | 100 | 500 | 500 |
| $49 | 200 | 200 | 500 | 500 |
| $48 | 0 | 300 | 0 | 700 |
| $47 | 0 | 300 | 0 | 900 |
| $46 | 0 | 300 | 0 | 1500 |
Size of existing or easily uncovered bids
| Market: | 1 | 2 | 3 | 4 |
|---|---|---|---|---|
| Bid price | Thin and Shallow | Thin but Deep | Broad but Shallow | Broad and Deep |
| $50 | 100 | 100 | 500 | 500 |
| $49 | 200 | 200 | 500 | 500 |
| $48 | 0 | 300 | 0 | 700 |
| $47 | 0 | 300 | 0 | 900 |
| $46 | 0 | 300 | 0 | 1500 |
Markets 3 and 4 are broader compared to markets 1 and 2 because larger volumes of orders (1000 units instead of 300) can be satisfied with no price impact ($49-50). Market 4 is also broader than market 2 because the price impact of larger volume orders occurs at greater volumes. In market 4, 1000 units can be sold at $49-50 whereas in market 2, the price concession goes as far as $46 to sell the same units.
Markets 2 and 4 are deeper compared to markets 1 and 3 because trade interests exist up to $46.
The examples also show that deep markets can compensate for broader markets. Thus, because market 2 is deep. 1200 units can be sold by dividing the sale into smaller amounts. In the broad but shallow market 3, only 1000 units can be sold.
continuous market. However, since dealers usually try to square their positions or maintain a specified structural position toward the end of the day, they only “provide” liquidity by taking inventory positions as long they assume buyers and sellers will continue to emerge.6
In a pure auction market, potential buyers and sellers submit orders, and brokers or an electronic system will match them. Auction markets are thus order or price driven and may be less continuous if there are few transactions. Market intermediaries in auction systems may also take inventory positions in order to facilitate liquidity (e.g., so-called specialists in widely traded securities). Most trading systems allow participants to submit limit-orders, which generally improve the liquidity. The intermediaries having direct access to the trading systems may cover their costs by charging a commission or they quote bid and ask prices to be paid by the ultimate buyers and sellers.
A distinction is also made between the primary market, where new issues are sold, and the secondary market, where those who have bought the issues at the primary market can resell them. The secondary market thus provides liquidity to those who have bought the securities.7 It is important to understand the reporting requirements of transactions in various markets before trading volumes can be used as a liquidity indicator.
While the paper focuses on measuring a financial market’s liquidity, it is important to note that the concept of liquidity is also used to discuss other types of liquidity. A distinction can be made between: (i) asset liquidity; (ii) an asset’s market liquidity; (iii) a financial market’s liquidity; and (iv) the liquidity of a financial institution. An asset is liquid if it can easily be converted into legal tender, which per definition is fully liquid. Some financial claims, like demand deposits, are virtually perfectly liquid—as long as the credit institution is liquid— since they can be converted without cost or delay into money during normal circumstances, while the transformation of other claims into legal tender may involve brokers’ commissions, settlement delays, etc. The emphasis here is on transaction costs and immediacy. The concept of an asset’s market liquidity is broader. It is related to the ease with which, in the absence of new information altering an asset’s fundamental price, large volumes of the asset can be disposed of quickly at a reasonable price.
A financial market’s liquidity depends on the substitutability among the various assets traded in a particular market, and how liquid each of these assets are. If there are different issuers, particularly in the corporate bond markets and equities markets, credit risk can prevent substitutability and result in significant segmentation of the market. In spite of having the same issuer, individual assets may still have different characteristics, such as different maturities in the market for government securities, different voting rights for preference shares, etc. This aggregation problem renders difficult an attempt to apply measures to individual assets with the objective of measuring a market’s liquidity.
Institutional liquidity, on the other hand, refers to how easily financial institutions can engage in financial transactions with a view to quickly cover mismatches between their assets and liabilities, which may be measured by liquid asset ratios, etc., and to settle their obligations.8 The more liquid the assets in its portfolio are and the less liquid the liabilities are, the greater the flexibility in managing asset-liability mismatches and its ability to meet settlement obligations.9 Financial institutions’ risk management systems increasingly rely on the assumption that their financial assets are liquid.
This paper identifies measures to gauge an asset’s market liquidity with a view to assess if a financial market, or at a minimum some of its segments, can be characterized as liquid. With a few exemptions, such as Chordia et al. (2001), who study market liquidity, and Chordia et al. (2002), who analyze the correlation of liquidity measures between markets, most studies have investigated the liquidity of individual financial assets rather than a financial market’s liquidity.
The rest of the paper is organized as follows. Section II classifies liquidity measures according to the dimension which they best measure. It also discusses factors that may affect their interpretation and ability to capture a given aspect of liquidity. Issues related to data availability to construct the measures are also briefly discussed. Section III applies the selected liquidity measures to the foreign exchange, money, and capital markets of a selected group of countries. Section IV lists some of the more important qualitative factors to consider when comparing the liquidity measures across markets and countries. Section V notes how liquidity measures during periods of stress may change. Section VI concludes the paper and briefly discusses how the measures presented could be used in the context of the FSAP.
II. Selected Liquidity Measures
Liquidity measures can be classified into four categories: (i) transaction cost measures that capture costs of trading financial assets and trading frictions in secondary markets; (ii) volume-based measures that distinguish liquid markets by the volume of transactions compared to the price variability, primarily to measure breadth and depth; (iii) equilibrium price-based measures that try to capture orderly movements towards equilibrium prices to mainly measure resiliency, and (iv) market-impact measures that attempt to differentiate between price movements due the degree of liquidity from other factors, such as general market conditions or arrival of new information to measure both elements of resiliency and speed of price discovery. No single measure, however, unequivocally measures tightness, immediacy, depth, breadth, and resiliency.
A. Transaction Cost Measures
A distinction can be made between explicit transaction costs, which relate to expenses such as order processing costs and taxes associated with trades, and implicit transaction (execution) costs. Because bid-ask spreads may capture nearly all of these costs, they are the most commonly used measure of transaction (execution) costs.
In dealer markets, the bid-ask spreads may reflect: (i) order-processing costs; (ii) asymmetric information costs; (iii) inventory-carrying costs; and (iv) oligopolistic market structure costs.10 Immediacy, for instance, is fostered by the existence of dealers who stand ready to buy and sell specific quantities of a financial instrument at the quoted bid and ask prices. This service entails inventory-carrying costs—depending on the dealers squaring their positions toward the end of the day—which they must recover in addition to their order processing costs. But dealers also incur a risk by standing ready to trade based on asymmetric information. They must charge a premium to compensate for potential losses in providing a continuous market. Such costs are smaller, if there are numerous participants willing to trade with the dealers, and thus revealing their asymmetric information. In addition, since immediacy is bought at a price, the latter is influenced by competition. Thus, a few dealers with oligopolistic power may have higher discretionary fees for immediacy.
High transaction costs reduce the demand for trades and therefore the number of potentially active participants in a market.11 This could also lead to more fragmented markets as many transactions may take place within the market makers’ spreads and not necessarily around the equilibrium price, which results in a shallow market. High spreads, or commissions in auction markets, also encourage transactors to seek potential counterparts in a trade outside the market makers’ markets; as such trades might be worth the search costs. That is, transactions will take place in the so-called fourth market. In contrast, when transactions costs are small, transactors would prefer to use dealers in auction mechanisms to trade rather than incur direct search costs, including through brokers. This results in transactions that are more likely to take place around the equilibrium price of an asset leading to a more unified and deep market.
The reduction in the number of market participants due to high transaction costs also affects breadth and resiliency. Since breadth implies having numerous participants, high transactions costs may lead to thin markets. Similarly, since large transaction costs may deter trades, they reduce resiliency by preventing orders from flowing in promptly to correct order imbalances that tend to move prices away from their fundamental level. In other words, the elasticity of order flows is generally much lower when transaction costs are high. The infrequency of trades is also likely to result in a market with substantial price discontinuities. The effects of high transactions costs go full circle, since a smaller number of participants reduce economies of scale on inventory costs with second round effects on market makers’ spreads.
The bid-ask spread can be measured as the absolute difference between bid and ask prices or as a percentage spread (equations 1.1 and 1.2 below). The percentage spread allows taking into account the fact that a given spread would be less costly the higher the prices, and it is easier to compare across markets. Dealers’ uncertainty about the equilibrium price also leads to adjustments in their bid and ask prices.
The bid-ask spread of the market is generally calculated using the highest bid and lowest ask prices in the market for a reference period, or in practice the most recent quotation. However, if there are several bid and ask prices available from different dealers and particularly if they are not obligated to trade at the quoted prices, consideration should be given to ignoring the most extreme outliers. This market spread should be distinguished from individual dealers’ spreads.
Other variants of the bid-ask spread can be calculated. Equations 1.1 and 1.2 above are calculated using quoted bid and ask prices. The bid-ask spread is sometimes calculated using weighted averages of actually executed trades over a period of time, since trades may not take place at quoted prices. In that case, the spread is called a realized spread.
In addition to the spread itself, the trade-size at which a dealer is committed to trade at quoted prices is also a useful indicator. All other things equal, the larger the trades that can be conducted at a quoted spread, the more depth and breadth the market has, but a large trade-size may also reduce the willingness to quote prices.12
Although not widespread, some of the high frequency databases now available occasionally include both bid and ask prices on a daily basis. However, both within the day, during the week and month, there may be patterns to take into account, thus monthly averages may not necessarily provide good indications of changes in the spread. Furthermore, on days when new important news are announced, the time it takes for the widened spread to reach “normal” levels can be an important indicator of resiliency.13 The financial asset with the shortest adjustment period is the most resilient.
B. Volume-Based Measures
Volume-based measures are most useful in measuring breadth (the existence of both numerous and large orders in volume with minimal transaction price impact). Markets that are deep tend to foster breadth since large orders can be divided into several smaller orders to minimize the impact on transaction prices.
Large numbers of trades are a valuable source of information for transactors and particularly dealers. They obtain information from order flows, and imbalances in this order flow give them information about the accuracy of their quoted prices. Changes in these quoted prices trigger balancing order flows, which would counter price movements that are not warranted by fundamentals (resiliency). This process allows dealers to have a continuous information source as to whether price changes are permanent or transitory. When markets lack breadth and depth, the absence of the continuous information source provided by numerous and frequent trades may result in price discontinuities and uncertainty about equilibrium prices. Even when there is uncertainty about equilibrium prices, but numerous orders from both the selling and buying sides of the market exist, transactors, and particularly dealers, may be able to execute orders without having to take risky inventory positions. Trading can also be enhanced if market makers can easily identify potential buyers and sellers, such as institutional investors with large portfolios.
Uncertainty about equilibrium prices may not, however, be a necessary outcome of a lack of breadth (orders are numerous and large in volume), and depth (wide range of orders), or higher transactions costs in a given market. Market participants may also infer equilibrium prices from the market of close substitutes, where price information is more complete. Thus, the existence of a deep and broad market for a close substitute may compensate for thinness, since they allow market makers to hedge position imbalances without waiting for balancing orders14. However, many players have learned the hard way that assets perceived as being close substitutes may not turn out to be so in practice.
Trading volume is traditionally used to measure the existence of numerous market participants and transactions. The trading volume can be given more meaning by relating it to the outstanding volume of the asset being considered. The resulting turnover rate (equations 2.1 and 2.2 below) gives an indication of the number of times the outstanding volume of the asset changes hands.
Pi and Qi are prices and quantities of the i trade during a specified period
V is as defined in (2.1).
S is the outstanding stock of the asset
P is the average price of the i trades in (2.1).
While it is relatively easy to estimate turnover rates in exchange traded securities markets, it is more difficult to choose an appropriate basis against which to measure turnover rates in the typical OTC foreign exchange and money markets. In the latter cases, provided data are available, the absolute trading volume and the number of transactions, and thus the average trade size, may be better measures of the existence of numerous and large trades, that is, dimensions of market breadth.
Finally, the trading volume may shift significantly both during the day, week, and month depending on trading patterns, for instance around announcement of new information important for the pertinent asset. The volatility of the turnover should thus also be taken into consideration.
The Hui-Heubel Liquidity Ratio (equation 2.3 below), originally applied to the equity market, attempts to capture the other dimension of market breadth, which relates the volumes of trades to their impact on prices, and thus also to resiliency.
The Lhh can be calculated as an average of the 5-day periods in a sample (e.g., 3 months) to smooth volatility. Subject to data availability, the ratio could also be calculated on a daily basis to capture very short-term price movements. The lower the LHH, the higher the liquidity of the asset. To be specific about the dimension of liquidity being captured, we would say that the market has more breadth when the LHH is low.
Pmax = highest daily price over last 5 days
Pmin = lowest daily price over last 5 days
V = total dollar volume traded last 5 days
S = number of instruments outstanding
= average closing price of the instrument over a 5-day period
The numerator in Lhh can simply be measured as the percentage change in the price of the asset over the 5 day period chosen. If those prices are not available, bid-ask prices could be used as a proxy to calculate the ratio, but then the information content also changes somewhat.
Conventional liquidity measures relate this price change to the simple volume traded in the denominator (V). The Hui-rieubel’s liquidity ratio uses in the denominator the ratio of the traded volume to the outstanding volume of the asset (essentially the turnover rate).15 Depending on data availability, other measures of trading volume can be used in the denominator (e.g., number of securities traded). Liquidity ratios in general can also be expressed in terms of the value or number of units traded in the numerator to the percent change for a given period interval. In this case, the market has more breadth, the larger the number of trades to the percentage price change.
It can be argued that the impact of trading a large volume of an asset on price depends on whether the volume traded is a high proportion of the volume of the asset held in the market, which the Hui-Heubel measure would capture. Thus, if buyers or sellers suddenly want to trade a high proportion of the outstanding volume of an asset, a significant price change could occur because those trades may be indications that new information arrived in the market. The price movement should therefore not be assimilated with illiquidity. As a result, one of the criticisms of these liquidity ratios is the fact that the relationship between price movements and volumes is not proportional. In using the ratio to predict future relationships between the two variables, one may overestimate price changes on large volumes and underestimate them on small volumes. Furthermore, there is no distinction between transitory and permanent price changes.
C. Price-Based Measures
Bernstein (1987) noted that “measures of liquidity when no information is hitting a stock must be more relevant than measures of liquidity when new information leads to new equilibrium values…thus unrefined measures of liquidity may be nothing more than some kind of weighted average reflecting the frequency with which new information hits one stock as compared with another.” Ideally, there is thus a need for an underlying structural model to identify the equilibrium price, but given the difficulty in determining whether new information is indeed affecting the price of an instrument, Hasbrouck and Schwartz (1988) proposed the market efficiency coefficient to distinguish short-term from long-term price changes.
The Market-Efficiency Coefficient (MEC) exploits the fact that price movements are more continuous in liquid markets, even if new information is affecting equilibrium prices (equation 3.1 below). Thus for a given permanent price change, the transitory changes to that price should be minimal in resilient markets.
Var (Rt) = variance of the logarithm of long-period returns16
Var (rt) = variance of the logarithm of short-period returns
T = number of short periods in each longer period
The ratio would tend to be closer but slightly below one in more resilient markets, since a minimum of short term volatility should be expected. Indeed, prices of assets with low market resiliency may exhibit greater volatility (more transitory changes) between periods in which their equilibrium price is changing. Factors that foster excessive short-period volatility (overshooting) result in an MEC substantially below one. These factors include price rounding, spreads, and inaccurate price discovery. On the other hand, Bernstein (1987, p. 12) notes that factors such as market maker intervention, and inaccurate price determination involving partial adjustment to news, cause prices to adjust in relatively small, and positively correlated increments. This would dampen short-period price volatility relative to longer-period price volatility, and may cause the MECs to be above one.
Low price volatility, when a new equilibrium is being established, is also related to the concept of orderly markets.17 Orderly and resilient markets provide for greater price continuity, which is a desirable feature of liquid markets. It should however be noted that discontinuity in price movements in order to reach a new equilibrium price warranted by new information is a feature of information efficient markets.18 The MEC should not render an unfavorable verdict on liquidity and resiliency if it is calculated over a given period in which the equilibrium price changed discretely in response to new information and then stabilized quickly.
Indeed, it can be questioned if price continuity is synonymous to resiliency. Recall that resiliency is a characteristic of markets in which new orders flow quickly to correct order imbalances that tend to move prices away from what is warranted by fundamentals. In practice there may ex ante be quite a range of views, including by the central bank, on what is warranted by fundamentals, which ex post may turn out to be quite different. A dealer or day-trader, for example, wanting to square her positions toward the end of the day has a different time horizon than a transactor having a medium-term horizon in mind. Whatever the case may be, if market participants are mostly on one side of the market because of new fundamentals, the resulting order imbalance should lead to a price change. If pressures for a price change are countered by new orders flowing in, the phenomenon may produce more price continuity, but this price continuity should not be associated with market resiliency. It would rather be associated with an orderly market and possibly an information inefficient market. In this case, the MEC may be greater than one (as noted above), because the volatility over the long period could be higher than that of the shorter periods.
It is important to note that a market can become one sided and lead to a significant price change although unjustified by fundamentals. In this case, even if market makers are able to determine the equilibrium price of an asset based on fundamentals, cash and regulatory constraints (e.g., leverage limits) may prevent them from absorbing an order imbalance without a significant price change. Resiliency is thus lost and liquidity evaporates in all its dimensions.19
The latter type of lost market liquidity should be distinguished from the one the paper has discussed so far. Wood and Wood (1985, p. 165-66) insightfully note that “even U.S. government securities are illiquid in the presence of widespread shortage of cash…General scramble for cash, or panic, leads to precipitate price falls....[in this case], nothing except cash is liquid.” They, like Keynes, also point out that liquidity perceived by an individual transactor is always greater than the liquidity if fully exploited by all transactors. If dealers are able to determine the equilibrium price of an asset, a single seller should not have difficulty selling that asset if she is acting more or less independently. On the other hand, if dealers are able to determine the correct price of an asset, but everyone else is selling, market liquidity is lacking. In the latter case, it is the perfectly liquid assets against which other assets’ liquidity are compared—namely cash—that is missing. Temporary injection of liquidity—cash—to support willingness to take open positions could in this case help market resiliency.20
Trading systems in which trading in a financial instrument is stopped when order imbalances are high (so-called circuit breakers) tend to reduce price continuity. Circuit breakers do so by allowing prices to move discretely after the pause in trading. The pause in trading may be needed because the orderly movement of prices associated with price continuity, may prevent a discrete price movement to a new equilibrium price (Bernstein, 1987). Many market commentators, however, have mixed views on the merit of circuit breakers. However, the halting of trades may be needed to foster fair markets by allowing all market participants to attempt to determine whether fundamentally new information has altered an asset’s equilibrium price. In this case, one would say that the market has temporarily lost price continuity. However, when trading resumes at a new price among informed traders, the market may still be a qualified as liquid by the definition given in the introduction of the paper. In other words, traders neutralizing small price deviations may no longer be around to help provide depth and breadth, but the market may still be liquid and resilient thanks to more informed traders, as the new information has already been absorbed. The MEC calculated over a long period covering a significant discrete price change may thus still be an appropriate measure of resiliency. However, it may not be true for all types of markets. Some argue that markets that are quote-driven generally provide more price continuity than markets that are order and call-driven, although it is debatable.
In addition to the MEC, vector auto regression econometric techniques, like impulse response functions, are also used to uncover the fact that the price discovery process is more timely and complete in liquid markets. Vector auto regression lags of price adjustments are shorter in liquid markets. As with other econometric techniques discussed below, operational ease argues against their use.
D. Market-Impact Measures
As noted above, liquidity ratios, such as the LHH, generally do not distinguish between transitory price changes from permanent ones warranted by new information. When new information becomes available in the market, even small transaction volumes could be associated with large price movements. For instance, new information triggering a financial crisis may not result in large turnovers because transactors, as long as they are not cash constrained, may prefer to wait and see. To better capture the price movement mainly due to large volumes, i.e. breadth, the price movements due to significant new information should ideally be extracted.
A distinction is often made in the equity markets between systematic and unsystematic risk based on the capital asset pricing model (CAPM), which also provides an avenue to extract market movements (equation 4.1 and 4.2 below). The systematic effect relates to a risk that cannot be diversified because it affects all securities in a systematic fashion. The degree of this effect is called the “beta of the stock” to refer to the regression coefficient of a stock’s return on that of the market. The higher the “beta,” the higher the systematic risk of that stock. The unsystematic risk is the risk that is specific to the stock in question, once the market risk is removed. Hui and Heubel, using this approach, suggested calculating the Market-Adjusted Liquidity for equities.
Ri = daily return on the ith stock
Rm = daily market return (e.g., S&P index)
β = regression coefficient, represents systematic risk
ui = regression residuals or specific risk.
The regression residual is then used to relate its variance to the volume traded:
= squared residuals from equation 4.1 Vi = daily percentage change in dollar volume traded
ej = equation 4.2 residuals.
The market-adjusted liquidity uses the residual of a regression of the asset’s return on the return of the market (thus purging it from its systematic risk) to determine the intrinsic liquidity of the asset.
The smaller γ2 in equation 4.2 (above), the smaller is the impact of trading volume on the variability of the assets’ price and therefore, the assets is more liquid. It should be noted that the lower the coefficient, the more breadth the market has. Note that the residuals of equation 4.1 could also have been used to calculate the LHH discussed in the previous section.
It is also possible to distinguish between the market impact, that is the change in the zero-coupon yield curve, and the liquidity premium of government bonds. The main differences among government bonds are typically their maturity and their type (bullet bonds, serial bonds, etc.). Zero-coupon yield curves can be estimated to, ideally, better take into account the different timing of interest and principal.21 Based on these yield curves, it is possible to estimate the liquidity premium of a particular bond, as the difference between the market price of the bond and the estimated price using the zero-coupon yield curve. In practice, however, the spread between benchmark government bond and a government security with roughly the same duration, but traded less, is often used as a proxy for the liquidity premium. In the case of corporate bonds, the spread between the corporate bond and the benchmark government security reflects both the difference in credit risk and a liquidity premium.
Newer research using high frequency data and a combination of macroeconomic models and microstructure models, like order flows (Evans and Lyons, 2002) or news impact (Melvin and Yin, 2000), reportedly do produce exchange rate forecasts outperforming random walks. The difference in part reflects the market’s liquidity (Galati and Ho, 2001). Foreign exchange markets are generally perceived as some of the most information efficient markets, in part because macroeconomic models using monthly data, rarely have outperformed random walk models.
Other econometric techniques
Other econometric techniques are used in some liquidity studies to separate the impact of anticipated trading volumes from those that are unanticipated and which may carry new information. The expected volumes are usually estimated by fitting an auto regressive moving average (ARMA) model of volumes traded. Actual volumes which deviate from the expected volumes as forecasted by the ARMA model are considered unexpected events, which are associated with new information flowing into the market. This distinction is used to explain the size of dealers’ spreads. Thus, high expected volumes of trades reduce the dealers’ spreads on account, for instance, of the economies of scale in their inventory costs discussed in section A. Unexpected volumes, however, will increase dealers’ spreads by increasing the uncertainty premium associated with their trading on potentially asymmetric information.
More sophisticated econometric techniques are also used to take account of the fact that once price volatility starts, it will take some time for all market participants to come to agreement on equilibrium prices. This results in volatility persistence, which can be captured by auto regressive conditional heteroskedasticity (ARCH) and generalized auto regressive conditional heteroskedasticity (GARCH) type models. These models simply say that a given period volatility is dependent on the volatilities of previous periods.22
Although more advanced econometric techniques have analytic appeal, they are not very operational. The computational burden may out-weigh the benefits. Operationally, to make a statement about market breadth, it may be easier to analyze trading volumes and price volatility patterns over a long period using simple liquidity ratios such as the Lhh and turnover figures. In doing so, one should keep in mind that the inferences that would be made regarding statistical relationships between price volatility and volume would be less precise, although general trends could be uncovered. Price-based measures, which were discussed in the previous subsection, attempt to make a statement about the degree of an asset’s liquidity by directly analyzing its price volatility. These measures avoid the issue of determining whether price movements are due to new information arriving in markets, thus they may actually be better measures of market resiliency.
III. Application of Liquidity Measures
This section analyzes various dimensions of liquidity in the foreign exchange, money, bond, and equity markets of a selected group of countries. While all measures cannot be applied in all markets because of lack of data (summarized in Box 2), several measures can be applied to compare the liquidity of different segments of a market, between markets, and between markets in different countries. Most of the data used in this section are publicly available information in the Bloomberg information system. The prices, however, are not firm, but merely indicative, and consequently we do not know how accurate they are. Additional data are typically available in central bank bulletins, publications issued by stock exchanges, dealer associations, etc.23 Nevertheless, when these liquidity measures are used in the context of an FSAP, it is often necessary to request additional data from the authorities, particularly to ensure access to daily observations and volume figures.24 Finally, there are several factors to keep in mind when applying the measures, as discussed in Section IV.
Liquidity Measures and Data Availability
In some securities markets, dealers account for a significant share of the turnover of listed securities. Bonds, for instance, are traded OTC in many markets.
| Measures | Typically OTC Markets | Typically Exchange Traded Markets1/ | ||
|---|---|---|---|---|
| Foreign exchange Markets | Money Markets | Bond Markets | Equity Markets | |
| Bid-ask spreads (mainly tightness supported by breadth and depth) | +++ data readily available | ++ data available for selected instruments and maturities | + data available for selected maturities or derivatives only | + data available for individual securities and index derivatives only |
| Turnover ratios (mainly breadth supported by depth and tightness) | + volume data rarely available, lack of denominator | + data available for selected maturities, lack of denominator | +++ data often available for the market and individual securities | +++ data often available for the market and individual securities |
| Market-Efficiency Coefficient (MEC) (mainly resiliency supported by faster price discovery and depth) | +++ data readily available | +++ data available for selected maturities | ++ data available for benchmarks and selected indexes | ++ data available for individual securities and market indexes |
In some securities markets, dealers account for a significant share of the turnover of listed securities. Bonds, for instance, are traded OTC in many markets.
| Measures | Typically OTC Markets | Typically Exchange Traded Markets1/ | ||
|---|---|---|---|---|
| Foreign exchange Markets | Money Markets | Bond Markets | Equity Markets | |
| Bid-ask spreads (mainly tightness supported by breadth and depth) | +++ data readily available | ++ data available for selected instruments and maturities | + data available for selected maturities or derivatives only | + data available for individual securities and index derivatives only |
| Turnover ratios (mainly breadth supported by depth and tightness) | + volume data rarely available, lack of denominator | + data available for selected maturities, lack of denominator | +++ data often available for the market and individual securities | +++ data often available for the market and individual securities |
| Market-Efficiency Coefficient (MEC) (mainly resiliency supported by faster price discovery and depth) | +++ data readily available | +++ data available for selected maturities | ++ data available for benchmarks and selected indexes | ++ data available for individual securities and market indexes |
In some securities markets, dealers account for a significant share of the turnover of listed securities. Bonds, for instance, are traded OTC in many markets.
A. Foreign Exchange Markets
Foreign exchange markets are generally perceived as some of the most liquid and information efficient markets, in part because it is a relatively homogeneous product and the daily turnover is significant. The Bank for International Settlements (BIS) conducts a survey in April every third year, and estimated that in 2001 the daily average foreign exchange turnover (i.e., spot, outright forwards, and foreign exchange swaps) in 48 countries, covering 2,772 banks, amounted to around US$1,210 billion, while the daily average turnover in OTC instruments amounted to US$67 billion (Figure 1). These figures are adjusted for double counting of local and cross-border interdealer transactions but include BIS’ estimations for reporting gaps. They are also adjusted for exchange rate developments during the period from 1989 to 2001. The decline in turnover from the survey in 1998 to 2001 should, according to the BIS, be seen in context of the introduction of the euro; the increased use of electronic broking, particularly in the spot markets; the consolidation in the banking industry; and, according to anecdotal information, the reduced activity of hedge funds. Most of the trading takes place vis-à-vis the U.S. dollar.25
Global Foreign Exchange Market Turnover, 1989–2001
Daily averages in April (billions of U.S. dollars)
Citation: IMF Working Papers 2002, 232; 10.5089/9781451875577.001.A001
Source: Bank for International Settlements, 2001
Global Foreign Exchange Market Turnover, 1989–2001
Daily averages in April (billions of U.S. dollars)
Citation: IMF Working Papers 2002, 232; 10.5089/9781451875577.001.A001
Source: Bank for International Settlements, 2001Global Foreign Exchange Market Turnover, 1989–2001
Daily averages in April (billions of U.S. dollars)
Citation: IMF Working Papers 2002, 232; 10.5089/9781451875577.001.A001
Source: Bank for International Settlements, 2001Except for such surveys, relatively few central banks do regularly publish information about the turnover in their respective foreign exchange markets. In addition, the lack of a proper base for the outstanding value of foreign exchange (K) prevents the calculation of liquidity ratios such as the Hui-Heubel ratio—(|%ΔP|/(V/K)).26 As a result, most liquidity measures for the foreign exchange markets focuses on bid-ask spreads (Tables 1 and 2) and exchange rate volatility used in the MEC.
Average Bid-Offer Spreads of Spot and Forwards, October 11, 20011/
Bear in mind that the spreads on one particular day of one dealer is only indicative and may deviate significantly from the average spreads over a longer period based on quotations from several dealers.
Average Bid-Offer Spreads of Spot and Forwards, October 11, 20011/
| Currency | Spot | 1-month | 2-month | 3-month | 6-month | 9-month | 1-year | |
|---|---|---|---|---|---|---|---|---|
| (percent of price, basis points) | ||||||||
| China | CNY | 0 | 3 | 4 | 4 | 6 | 7 | 8 |
| Hong Kong SAR | HKD | 1 | 2 | 2 | 2 | 2 | 3 | 3 |
| Indonesia | IDR | 43 | 130 | 139 | 164 | 231 | 303 | 371 |
| India | INR | 4 | 27 | 30 | 30 | 40 | 54 | 67 |
| Korea | KRW | 3 | 9 | 9 | 9 | 9 | 9 | 9 |
| Philippines | PHP | 8 | 22 | 24 | 25 | 34 | 46 | 57 |
| Singapore | SGD | 5 | 8 | 8 | 9 | 10 | 12 | 14 |
| Thailand | THB | 16 | 26 | 32 | 35 | 42 | 57 | 77 |
| Taiwan Province of China | TWD | 3 | 10 | 12 | 12 | 14 | 15 | 16 |
| Argentina | ARS | 2 | 35 | 55 | 59 | 300 | 329 | 181 |
| Brazil | BRL | 6 | 20 | 29 | 36 | 40 | 46 | 44 |
| Chile | CLP | 5 | 9 | 13 | 18 | 25 | 36 | 47 |
| Colombia | COP | 7 | 25 | 35 | 47 | 83 | 104 | 121 |
| Mexico | MXN | 14 | 16 | 17 | 18 | 23 | 22 | 24 |
| Peru | PEN | 2 | 13 | 24 | 31 | 55 | 89 | 101 |
| Venezuela | VEB | 13 | 73 | 135 | 182 | 261 | 360 | 429 |
| Czech Republic | CZK | 5 | 7 | 7 | 7 | 9 | 10 | 11 |
| Hungary | HUF | 2 | 6 | 10 | 13 | 20 | 28 | 32 |
| Poland | PLN | 8 | 10 | 11 | 12 | 14 | 17 | 17 |
| Slovakia | SKK | 5 | 8 | 10 | 12 | 19 | 25 | 32 |
| Israel | ILS | 27 | 29 | 31 | 32 | 36 | 43 | 43 |
| Saudi Arabia | SAR | 2 | 3 | 4 | 5 | 6 | 8 | 8 |
| Turkey | TRL | 86 | 289 | 467 | 621 | 988 | 1255 | 1447 |
| South Africa | ZAR | 10 | 13 | 14 | 15 | 20 | 21 | 24 |
Bear in mind that the spreads on one particular day of one dealer is only indicative and may deviate significantly from the average spreads over a longer period based on quotations from several dealers.
Average Bid-Offer Spreads of Spot and Forwards, October 11, 20011/
| Currency | Spot | 1-month | 2-month | 3-month | 6-month | 9-month | 1-year | |
|---|---|---|---|---|---|---|---|---|
| (percent of price, basis points) | ||||||||
| China | CNY | 0 | 3 | 4 | 4 | 6 | 7 | 8 |
| Hong Kong SAR | HKD | 1 | 2 | 2 | 2 | 2 | 3 | 3 |
| Indonesia | IDR | 43 | 130 | 139 | 164 | 231 | 303 | 371 |
| India | INR | 4 | 27 | 30 | 30 | 40 | 54 | 67 |
| Korea | KRW | 3 | 9 | 9 | 9 | 9 | 9 | 9 |
| Philippines | PHP | 8 | 22 | 24 | 25 | 34 | 46 | 57 |
| Singapore | SGD | 5 | 8 | 8 | 9 | 10 | 12 | 14 |
| Thailand | THB | 16 | 26 | 32 | 35 | 42 | 57 | 77 |
| Taiwan Province of China | TWD | 3 | 10 | 12 | 12 | 14 | 15 | 16 |
| Argentina | ARS | 2 | 35 | 55 | 59 | 300 | 329 | 181 |
| Brazil | BRL | 6 | 20 | 29 | 36 | 40 | 46 | 44 |
| Chile | CLP | 5 | 9 | 13 | 18 | 25 | 36 | 47 |
| Colombia | COP | 7 | 25 | 35 | 47 | 83 | 104 | 121 |
| Mexico | MXN | 14 | 16 | 17 | 18 | 23 | 22 | 24 |
| Peru | PEN | 2 | 13 | 24 | 31 | 55 | 89 | 101 |
| Venezuela | VEB | 13 | 73 | 135 | 182 | 261 | 360 | 429 |
| Czech Republic | CZK | 5 | 7 | 7 | 7 | 9 | 10 | 11 |
| Hungary | HUF | 2 | 6 | 10 | 13 | 20 | 28 | 32 |
| Poland | PLN | 8 | 10 | 11 | 12 | 14 | 17 | 17 |
| Slovakia | SKK | 5 | 8 | 10 | 12 | 19 | 25 | 32 |
| Israel | ILS | 27 | 29 | 31 | 32 | 36 | 43 | 43 |
| Saudi Arabia | SAR | 2 | 3 | 4 | 5 | 6 | 8 | 8 |
| Turkey | TRL | 86 | 289 | 467 | 621 | 988 | 1255 | 1447 |
| South Africa | ZAR | 10 | 13 | 14 | 15 | 20 | 21 | 24 |
Bear in mind that the spreads on one particular day of one dealer is only indicative and may deviate significantly from the average spreads over a longer period based on quotations from several dealers.
Bid-Ask Spreads (basis points1/): Foreign Exchange Markets
Unless otherwise indicated
Bid-Ask Spreads (basis points1/): Foreign Exchange Markets
| 1996 | 1997 | 1998 | 1999 | 2000 | ||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | 02 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | 03 | Q4 | Avg | Q1 | 02 | Q3 | Q4 | ||
| Canada | 0.01 | 0.02 | 0.01 | 0 01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | |
| Australia | 0.07 | 0.07 | 0.06 | 0.06 | 0.07 | 0.07 | 0.07 | 0.06 | 0.08 | 0.08 | 0.09 | 0.10 | 0.09 | 0.08 | 0.09 | 0.08 | 0.08 | 0.07 | 0.08 | 0.09 | 0.09 | 0.08 | 0.09 | 0.09 | 0.11 | |
| Mexico | 0.23 | 0.34 | 0.27 | 0.20 | 0.14 | 0.18 | 0.15 | 0.13 | 0.13 | 0.32 | 0.22 | 0.22 | 0.28 | 0.23 | 0.14 | 0.16 | 0.21 | 0.17 | 0.14 | 0.13 | 0.13 | 0.13 | 0.12 | 0.16 | 0.11 | |
| Korea | 0.03 | 0.00 | 0.00 | 0.00 | 0.11 | -0.02 | 0.00 | 0.00 | 0.00 | -0.08 | 0.04 | 0.07 | 0.06 | 0.02 | 0.00 | 0.10 | 0.07 | 0.08 | 0.18 | 0.07 | 0.19 | 0.39 | 0.17 | 0.04 | 0.17 | |
| Singapore | 0.09 | 0.03 | 0.09 | 0.08 | 0.09 | 0.07 | 0.07 | 0.06 | 0.08 | 0.08 | 0.12 | 0.16 | 0.13 | 0.10 | 0.09 | 0.05 | 0.06 | 0.05 | 0.05 | 0.05 | 0.04 | 0.04 | 0.06 | 0.04 | 0.04 | |
| Malaysia | 0.06 | 0.08 | 0.05 | 0.05 | 0.06 | 0.09 | 0.04 | 0.04 | 0.11 | 0.15 | 0.36 | 0.49 | 0.40 | 0.46 | 0.07 | 0.08 | 0.11 | 0.04 | 0.04 | 0.13 | 0.03 | 0.05 | 0.02 | 0.02 | 0.02 | |
| South Africa | 0.23 | 0.20 | 0.37 | 0.14 | 0.24 | 0.12 | 0.18 | 0.11 | 0.09 | 0.09 | 0.19 | 0.08 | 0.10 | 0.27 | 0.30 | 0.12 | 0.20 | 0.11 | 0.10 | 0.08 | 0.10 | 0.08 | 0.11 | 0.09 | 0.10 | |
| Indonesia | -0.03 | 0.01 | 0.01 | -0.17 | 0.01 | 0.13 | 0.04 | 0.07 | 0.18 | 0.23 | 2.38 | 2.85 | 2.26 | 2.49 | 1.93 | 0.86 | 1.07 | 0.90 | 0.73 | 0.73 | 0.53 | 0.49 | 0.54 | 0.57 | 0.54 | |
| %ΛP | 0.10 | 0.12 | 0.07 | 0.14 | 0.06 | 1.08 | 0.08 | 0.10 | 1.17 | 2.92 | 3.04 | 5.02 | 3.06 | 2.00 | 2.14 | 1.29 | 1.48 | 1.05 | 1.50 | 1.15 | 0.79 | 0.73 | 0.75 | 1.05 | 0.64 | |
| V | … | … | … | … | … | 7.7E+03 | 8.3E+Q3 | 9.1E+03 | 8.0E+03 | 5.2E+03 | 1.3E+03 | 1.8E+03 | 1.4E+03 | 1.1E+03 | 8.8E+02 | 1.1E+03 | 8.8E+02 | 1.2E+03 | 1.3E+03 | 1.1E+03 | 1.4E+03 | 1.2E+03 | 1.6E+03 | 1.4E+03 | … | |
| %ΔP/V | … | … | … | … | … | 7.6E-04 | 1.2E-05 | 1.0E-04 | 2.2E-04 | 2.7E-03 | 2.1E-03 | 2.3E-03 | 1.8E-03 | 2.5E-03 | 1.8E-03 | 9.9E-04 | 9.0E-04 | 1.1E-03 | 1.2E-03 | 7.3E-04 | 5.7E-04 | 4.8E-04 | 5.7E-04 | 8.5E-04 | … | |
Unless otherwise indicated
Bid-Ask Spreads (basis points1/): Foreign Exchange Markets
| 1996 | 1997 | 1998 | 1999 | 2000 | ||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | 02 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | 03 | Q4 | Avg | Q1 | 02 | Q3 | Q4 | ||
| Canada | 0.01 | 0.02 | 0.01 | 0 01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | |
| Australia | 0.07 | 0.07 | 0.06 | 0.06 | 0.07 | 0.07 | 0.07 | 0.06 | 0.08 | 0.08 | 0.09 | 0.10 | 0.09 | 0.08 | 0.09 | 0.08 | 0.08 | 0.07 | 0.08 | 0.09 | 0.09 | 0.08 | 0.09 | 0.09 | 0.11 | |
| Mexico | 0.23 | 0.34 | 0.27 | 0.20 | 0.14 | 0.18 | 0.15 | 0.13 | 0.13 | 0.32 | 0.22 | 0.22 | 0.28 | 0.23 | 0.14 | 0.16 | 0.21 | 0.17 | 0.14 | 0.13 | 0.13 | 0.13 | 0.12 | 0.16 | 0.11 | |
| Korea | 0.03 | 0.00 | 0.00 | 0.00 | 0.11 | -0.02 | 0.00 | 0.00 | 0.00 | -0.08 | 0.04 | 0.07 | 0.06 | 0.02 | 0.00 | 0.10 | 0.07 | 0.08 | 0.18 | 0.07 | 0.19 | 0.39 | 0.17 | 0.04 | 0.17 | |
| Singapore | 0.09 | 0.03 | 0.09 | 0.08 | 0.09 | 0.07 | 0.07 | 0.06 | 0.08 | 0.08 | 0.12 | 0.16 | 0.13 | 0.10 | 0.09 | 0.05 | 0.06 | 0.05 | 0.05 | 0.05 | 0.04 | 0.04 | 0.06 | 0.04 | 0.04 | |
| Malaysia | 0.06 | 0.08 | 0.05 | 0.05 | 0.06 | 0.09 | 0.04 | 0.04 | 0.11 | 0.15 | 0.36 | 0.49 | 0.40 | 0.46 | 0.07 | 0.08 | 0.11 | 0.04 | 0.04 | 0.13 | 0.03 | 0.05 | 0.02 | 0.02 | 0.02 | |
| South Africa | 0.23 | 0.20 | 0.37 | 0.14 | 0.24 | 0.12 | 0.18 | 0.11 | 0.09 | 0.09 | 0.19 | 0.08 | 0.10 | 0.27 | 0.30 | 0.12 | 0.20 | 0.11 | 0.10 | 0.08 | 0.10 | 0.08 | 0.11 | 0.09 | 0.10 | |
| Indonesia | -0.03 | 0.01 | 0.01 | -0.17 | 0.01 | 0.13 | 0.04 | 0.07 | 0.18 | 0.23 | 2.38 | 2.85 | 2.26 | 2.49 | 1.93 | 0.86 | 1.07 | 0.90 | 0.73 | 0.73 | 0.53 | 0.49 | 0.54 | 0.57 | 0.54 | |
| %ΛP | 0.10 | 0.12 | 0.07 | 0.14 | 0.06 | 1.08 | 0.08 | 0.10 | 1.17 | 2.92 | 3.04 | 5.02 | 3.06 | 2.00 | 2.14 | 1.29 | 1.48 | 1.05 | 1.50 | 1.15 | 0.79 | 0.73 | 0.75 | 1.05 | 0.64 | |
| V | … | … | … | … | … | 7.7E+03 | 8.3E+Q3 | 9.1E+03 | 8.0E+03 | 5.2E+03 | 1.3E+03 | 1.8E+03 | 1.4E+03 | 1.1E+03 | 8.8E+02 | 1.1E+03 | 8.8E+02 | 1.2E+03 | 1.3E+03 | 1.1E+03 | 1.4E+03 | 1.2E+03 | 1.6E+03 | 1.4E+03 | … | |
| %ΔP/V | … | … | … | … | … | 7.6E-04 | 1.2E-05 | 1.0E-04 | 2.2E-04 | 2.7E-03 | 2.1E-03 | 2.3E-03 | 1.8E-03 | 2.5E-03 | 1.8E-03 | 9.9E-04 | 9.0E-04 | 1.1E-03 | 1.2E-03 | 7.3E-04 | 5.7E-04 | 4.8E-04 | 5.7E-04 | 8.5E-04 | … | |
Unless otherwise indicated
Table 2 shows the bid-ask spreads in basis points for a selected group of countries during the period 1996–2000. The data shows that Canada has the market with the lowest transactions costs as measured by the bid-ask spreads. The spread has also remained fairly unchanged throughout the sample period. In comparison to the other countries, the Canadian foreign exchange market with lower transaction costs can thus be characterized as more favorable to market depth. This, however, depends on the degree of the accuracy of the information source. Accordingly, country and market comparisons can be misleading.
Other countries show more variability in their bid-ask spreads during the sample period. There also seem to be some correlation around the Asian crises in late 1997 and the Russian crisis in August 1998. Countries such as Mexico and South Africa show declining spreads since 1998. Korea, on the other hand, shows deterioration after 1998. This deterioration probably reflects the effect of exchange rate flexibility since foreign exchange risk is part of the transaction costs implicit in the bid-ask spreads. This also helps explain the increase in bid-ask spreads in Malaysia during the 1998 period (Figure 1 in Appendix II).
However, while Malaysia’s bid-ask spreads have decreased with the return to an exchange rate peg, the MEC values in Table 3 show that resiliency seems to have deteriorated.27 This deterioration is consistent with the fact that the lower bid-ask spreads have become more volatile in the short run.
Market Efficiency Coefficients: Foreign Exchange Markets
Market Efficiency Coefficients: Foreign Exchange Markets
| 1996 | 1997 | 1998 | 1999 | 2000 | |||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | |
| Canada | 0.74 (0.27) | 0.77 | 0.65 | 0.66 | 0.91 | 0.90 (0.31) | 0.92 | 0.89 | 0.92 | 0.86 | 0.86 (0 34) | 1.05 | 0 69 | 0.78 | 0.92 | 0.91 (0.36) | 1.27 | 0.82 | 0.62 | 0.94 | 0.86 (0.29) | 0.70 | 1.07 | 0.98 | 0.67 |
| Australia | 0.77 (0.23) | 0.71 | 0.66 | 0.87 | 0.81 | 0.73 (0.27) | 0.67 | 0.63 | 0.84 | 0.76 | 0.79 (0.32) | 0.61 | 0.50 | 1.10 | 0.92 | 0.99 (0 32) | 1.44 | 0.96 | 0.84 | 0.75 | 0.71 (0.23) | 0.71 | 0.64 | 0.87 | 0.64 |
| Mexico | 0.95 (0.39) | 0.80 | 1.04 | 0.93 | 0.99 | 0.70 (0.35) | 0.73 | 0.92 | 0 52 | 0 62 | 0.63 (0.18) | 0.63 | 0.54 | 0.56 | 0.77 | 0.99 (0.35) | 0.91 | 1.18 | 0.72 | 1.14 | 0.84 (0.38) | 0.74 | 0.85 | 1.05 | 0.72 |
| Korea | 0.86 (0.41) | 0.93 | 0.91 | 0.78 | 0 85 | 0.50 (0 34) | 0.73 | 0.44 | 0.22 | 0.61 | 0.94 (0.28) | 0.80 | 0.96 | 1.07 | 0.93 | 0.88 (0.36) | 0.98 | 0.83 | 0.87 | 0.84 | 0.81 (0.45) | 0.54 | 0.92 | 0.68 | 1.08 |
| Singapore | 0.64 (0.25) | 0.30 | 0.91 | 0.66 | 0.59 | 0.63 (0 22) | 0.51 | 0.66 | 0.76 | 0.60 | 0.88 (0.28) | 0.84 | 0.99 | 0.71 | 0.97 | 0.74 (0.33) | 0.53 | 0.97 | 0.54 | 0.91 | 1.01 (0.32) | 0.77 | 1.15 | 1.07 | 1.05 |
| Malaysia | 0.71 (0.28) | 0.43 | 0.67 | 0.67 | 0.96 | 0.95 (0.41) | 0.83 | 0.70 | 1.02 | 1.25 | 0.81 (1.08) | 1.10 | 0.68 | 0.69 | 0.77 | 0.35 (0.61) | 0.22 | 0.35 | 0.50 | 0.31 | 0.22 (0 06) | 0.19 | 0.23 | 0.18 | 0.28 |
| Indonesia | 0.71 (0.62) | 0 61 | 0.65 | 0.49 | 1.04 | 0.88 (0.43) | 0.54 | 0.70 | 1.02 | 1.22 | 1.03 (0.38) | 0.89 | 0.95 | 1.14 | 1.15 | 0.91 (0.34) | 0.64 | 0.81 | 0.99 | 1.20 | 0.74 (0.32) | 0.63 | 0.98 | 0.88 | 0.48 |
| South Africa | 0.86 (0.36) | 1.13 | 0.64 | 0.86 | 0.89 | 0.85 (0.26) | 0.92 | 0.77 | 0.82 | 0.90 | 0.82 (0.27) | 0.93 | 0.79 | 0.63 | 0.95 | 0.85 (0.27) | 0.69 | 0.98 | 0.86 | 0.88 | 0.80 (0.22) | 0.84 | 0.74 | 0.76 | 0.85 |
Market Efficiency Coefficients: Foreign Exchange Markets
| 1996 | 1997 | 1998 | 1999 | 2000 | |||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | |
| Canada | 0.74 (0.27) | 0.77 | 0.65 | 0.66 | 0.91 | 0.90 (0.31) | 0.92 | 0.89 | 0.92 | 0.86 | 0.86 (0 34) | 1.05 | 0 69 | 0.78 | 0.92 | 0.91 (0.36) | 1.27 | 0.82 | 0.62 | 0.94 | 0.86 (0.29) | 0.70 | 1.07 | 0.98 | 0.67 |
| Australia | 0.77 (0.23) | 0.71 | 0.66 | 0.87 | 0.81 | 0.73 (0.27) | 0.67 | 0.63 | 0.84 | 0.76 | 0.79 (0.32) | 0.61 | 0.50 | 1.10 | 0.92 | 0.99 (0 32) | 1.44 | 0.96 | 0.84 | 0.75 | 0.71 (0.23) | 0.71 | 0.64 | 0.87 | 0.64 |
| Mexico | 0.95 (0.39) | 0.80 | 1.04 | 0.93 | 0.99 | 0.70 (0.35) | 0.73 | 0.92 | 0 52 | 0 62 | 0.63 (0.18) | 0.63 | 0.54 | 0.56 | 0.77 | 0.99 (0.35) | 0.91 | 1.18 | 0.72 | 1.14 | 0.84 (0.38) | 0.74 | 0.85 | 1.05 | 0.72 |
| Korea | 0.86 (0.41) | 0.93 | 0.91 | 0.78 | 0 85 | 0.50 (0 34) | 0.73 | 0.44 | 0.22 | 0.61 | 0.94 (0.28) | 0.80 | 0.96 | 1.07 | 0.93 | 0.88 (0.36) | 0.98 | 0.83 | 0.87 | 0.84 | 0.81 (0.45) | 0.54 | 0.92 | 0.68 | 1.08 |
| Singapore | 0.64 (0.25) | 0.30 | 0.91 | 0.66 | 0.59 | 0.63 (0 22) | 0.51 | 0.66 | 0.76 | 0.60 | 0.88 (0.28) | 0.84 | 0.99 | 0.71 | 0.97 | 0.74 (0.33) | 0.53 | 0.97 | 0.54 | 0.91 | 1.01 (0.32) | 0.77 | 1.15 | 1.07 | 1.05 |
| Malaysia | 0.71 (0.28) | 0.43 | 0.67 | 0.67 | 0.96 | 0.95 (0.41) | 0.83 | 0.70 | 1.02 | 1.25 | 0.81 (1.08) | 1.10 | 0.68 | 0.69 | 0.77 | 0.35 (0.61) | 0.22 | 0.35 | 0.50 | 0.31 | 0.22 (0 06) | 0.19 | 0.23 | 0.18 | 0.28 |
| Indonesia | 0.71 (0.62) | 0 61 | 0.65 | 0.49 | 1.04 | 0.88 (0.43) | 0.54 | 0.70 | 1.02 | 1.22 | 1.03 (0.38) | 0.89 | 0.95 | 1.14 | 1.15 | 0.91 (0.34) | 0.64 | 0.81 | 0.99 | 1.20 | 0.74 (0.32) | 0.63 | 0.98 | 0.88 | 0.48 |
| South Africa | 0.86 (0.36) | 1.13 | 0.64 | 0.86 | 0.89 | 0.85 (0.26) | 0.92 | 0.77 | 0.82 | 0.90 | 0.82 (0.27) | 0.93 | 0.79 | 0.63 | 0.95 | 0.85 (0.27) | 0.69 | 0.98 | 0.86 | 0.88 | 0.80 (0.22) | 0.84 | 0.74 | 0.76 | 0.85 |
The MEC figures in Table 3 are mostly below one, as expected, with some cases of marked deterioration in resiliency such as Korea during 1997. The MEC for Korea averaged 0.50 in 1997, when the exchange rate moved from W 900 per US$1 in August 1997 to over Wl, 600 per US$1 by December 1997. This period was preceded by significant decreases in resiliency from the first to the third quarter of 1997, when the MEC reached its lowest value of 0.22 from 0.85 in the last quarter of 1996. The very low MEC suggests inadequate price discovery during this period leading to excessive short-term exchange rate volatility. The excessive short-term volatility, in turn, may reflect the lack of resiliency or orders quickly flowing in to correct imbalances that tend to move prices away from equilibrium, in part due to uncertainty regarding fundamentals.
Indonesia also experienced a sharp exchange rate adjustment in late 1997, but in contrast to Korea, the MEC values increased throughout 1997, from 0.54 in the first quarter to 1.22 in the last quarter. As noted earlier, large MEC values reflect dampening effects on short-term price movements (e.g., foreign exchange market intervention, or inaccurate price determination), which lead to correlated but low short-term price volatility. This results in longer-term volatility being larger than short-term volatility and thus MEC values larger than one. A large MEC may therefore be a leading indicator of an adjustment in the equilibrium price gradually taking place.
In sum, during the financial crises in 1997, in both Korea and Indonesia, significantly low or high MEC values seem to have preceded large exchange rate adjustments. This pattern can be contrasted with that observed in South Africa, where MEC values have consistently been around 0.80 and the exchange rate seems to have depreciated smoothly since 1995, which may in part reflect a freely floating exchange rate policy (Figure 1, Appendix II).
In Indonesia, the volume data of monthly foreign exchange transactions show that the value of foreign exchange transactions (V) has greatly decreased since 1997 (Figure 5 in Appendix II), but the conventional liquidity ratio (|%ΔP|/V) has improved significantly over the same period. This in part reflects reduced exchange rate volatility from 1997 to 2000 (|%ΔP| and Figure 2 for Indonesia in Appendix II). This evidence suggests increased depth, which also is suggested by the bid-ask spreads that have decreased steadily from 1997 to 2000 (Figure 4 for Indonesia in Appendix II).
B. Money Markets
The money market consists of a number of different financial instruments with maturities up to one year. They typically include: (i) unsecured deposits/loans, which may be affected by credit risk; (ii) secured deposits/loans in form of repurchase agreements (where ownership changes) or with a collateral agreement (pledging); (iii) foreign exchange swaps; (iv) short-term central bank bills; (v) short-term government securities (treasury bills); and (vi) commercial paper. Derivatives, such as forward rate agreements (FRA), futures, and options may also be traded in the money market. Most of these instruments are standardized, but nonstandard instruments may be traded as well, like foreign exchange swaps with unusual maturities, etc. The central bank typically intervenes in the most “liquid” segments of the money market with a view to influence the liquidity—reserve—conditions in the banking system or to observe an interest rate target in countries with well-developed financial markets and high degree of capital mobility. In addition to different instruments, there may also be different markets, for instance an electronic organized money market and an OTC market. Although arbitrage should reduce price differences among the various segments and markets to primarily reflect credit risk and the maturity structure, there may be significant market frictions.
Turnover figures are important, since they provide information about the composition of the money market and thus which segments better reveal the degree of liquidity in the money market. Central banks typically collect such information, but often daily information is not published. In small open economies without restrictions on capital movements, the foreign exchange market can indirectly be an important part of the money market. In the case of Denmark, for instance, foreign exchange swaps account for about 45 percent of the money market transactions. In addition to the absolute turnover, the volatility of the turnover is also important. It reflects a number of country specific issues, like averaging of required reserves, the location of the central government’s deposits, and the functioning of the payment system.28 In many countries, however, only money market rates and occasionally bid-ask spreads are readily available.
Table 4 shows quarterly averages of bid-ask spreads in Singapore, and Poland for overnight, 1-month, and 3-month maturities, respectively. The bid-ask spread varies from around 10 basis points to more than 100 basis points during periods of uncertainty, such as the fourth quarter of 1997 (the Asian crisis) and third quarter of 1998 (the Russian crisis), with Singapore relatively more affected by the Asian crisis than Poland by the Russian crisis. The volatility of the bid-ask spreads, which may be better inferred from the figures in Appendix III, shows that in Poland, the spread in selected segments of the interbank market is rather volatile, while in Singapore, the spreads have remained fairly constant after the effects of the Asian crises in 1997–98 were worn out.
Bid-Ask Spreads: Money Markets
Bid-Ask Spreads: Money Markets
| Maturity | 1997 | 1998 | 1999 | 2000 | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | 04 | ||
| Singapore | Overnight | 0.33 (1.00) | 0.13 | 0.12 | 0.16 | 0.87 | 1.02 (1.36) | 1.38 | 1.09 | 1.00 | 0.62 | 0.24 (0.11) | 0.26 | 0.24 | 0.25 | 0.22 | 0.25 (0.04) | 0.24 | 0.26 | 0.25 | 0.25 |
| 1 Month | 0.31 (0.43) | 0.13 | 0.13 | 0.16 | 0.81 | 0.86 (0.90) | 1.45 | 1.04 | 0.55 | 0.47 | 0.26 (0.05) | 0.30 | 0.25 | 0.24 | 0.26 | 0.25 (0.02) | 0.25 | 0.25 | 0.25 | 0.24 | |
| 3 Month | 0.28 (0.38) | 0.13 | 0.12 | 0.14 | 0.73 | 0.86 (0.63) | 1.42 | 0.95 | 0.64 | 0.47 | 0.29 (0.07) | 0.38 | 0.29 | 0.24 | 0.25 | 0.25 (0.02) | 0.25 | 0.25 | 0.25 | 0.24 | |
| Poland | Overnight | 0.93 (1.86) | 0.46 | 0.84 | 1.09 | 1.32 | 0.64 (0.79) | 0.70 | 0.50 | 0.67 | 0.69 | 0.70 (1.25) | 0.35 | 0.20 | 0.37 | 1.92 | 1.21 (1.52) | 1.49 | 1.26 | 1.26 | 0.86 |
| 1 Month | 0.84 (0.35) | 0.62 | 0.61 | 1.12 | 0.98 | 0.73 (0.35) | 1.06 | 0.62 | 0.71 | 0.50 | 0.58 (0.84) | 0.33 | 0.17 | 0.20 | 1.66 | 0.42 (0.21) | 0.54 | 0.34 | 0.34 | 0.43 | |
| 3 Month | 0.98 (0.38) | 0.74 | 0.65 | 1.37 | 1.14 | 1.02 (0.30) | 1.15 | 0.84 | 1.01 | 1.05 | 0.80 (0.73) | 0.64 | 0.28 | 0.42 | 1.92 | 0.53 (0.25) | 0.78 | 0.52 | 0.44 | 0.39 | |
Bid-Ask Spreads: Money Markets
| Maturity | 1997 | 1998 | 1999 | 2000 | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | 04 | ||
| Singapore | Overnight | 0.33 (1.00) | 0.13 | 0.12 | 0.16 | 0.87 | 1.02 (1.36) | 1.38 | 1.09 | 1.00 | 0.62 | 0.24 (0.11) | 0.26 | 0.24 | 0.25 | 0.22 | 0.25 (0.04) | 0.24 | 0.26 | 0.25 | 0.25 |
| 1 Month | 0.31 (0.43) | 0.13 | 0.13 | 0.16 | 0.81 | 0.86 (0.90) | 1.45 | 1.04 | 0.55 | 0.47 | 0.26 (0.05) | 0.30 | 0.25 | 0.24 | 0.26 | 0.25 (0.02) | 0.25 | 0.25 | 0.25 | 0.24 | |
| 3 Month | 0.28 (0.38) | 0.13 | 0.12 | 0.14 | 0.73 | 0.86 (0.63) | 1.42 | 0.95 | 0.64 | 0.47 | 0.29 (0.07) | 0.38 | 0.29 | 0.24 | 0.25 | 0.25 (0.02) | 0.25 | 0.25 | 0.25 | 0.24 | |
| Poland | Overnight | 0.93 (1.86) | 0.46 | 0.84 | 1.09 | 1.32 | 0.64 (0.79) | 0.70 | 0.50 | 0.67 | 0.69 | 0.70 (1.25) | 0.35 | 0.20 | 0.37 | 1.92 | 1.21 (1.52) | 1.49 | 1.26 | 1.26 | 0.86 |
| 1 Month | 0.84 (0.35) | 0.62 | 0.61 | 1.12 | 0.98 | 0.73 (0.35) | 1.06 | 0.62 | 0.71 | 0.50 | 0.58 (0.84) | 0.33 | 0.17 | 0.20 | 1.66 | 0.42 (0.21) | 0.54 | 0.34 | 0.34 | 0.43 | |
| 3 Month | 0.98 (0.38) | 0.74 | 0.65 | 1.37 | 1.14 | 1.02 (0.30) | 1.15 | 0.84 | 1.01 | 1.05 | 0.80 (0.73) | 0.64 | 0.28 | 0.42 | 1.92 | 0.53 (0.25) | 0.78 | 0.52 | 0.44 | 0.39 | |
In the United States, where the Federal Reserve Bank targets the Fed funds rate, the MEC shows that the long period Fed funds returns (5-day return) is less volatile than the daily returns, thus showing a low MEC (Table 5). According to Furfine (2001), who calculates intraday volatility, 76 percent of the change takes place during the day. The MECs are higher for longer maturities; but on average they are below one and they are generally also more
Market Efficiency Coefficients: Money Markets
Market Efficiency Coefficients: Money Markets
| Maturity | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | |||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Ql | Q2 | Q3 | 04 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | ||
| U.S | Overnight | 0.30 (0.34) | 0.50 | 0.25 | 0.30 | 0.25 | 0.29 (0.18) | 0.37 | 0 24 | 0.26 | 0.31 | 0.29 (0.15) | 0.29 | 0.20 | 0.34 | 0.31 | 0.31 (0.11) | 0.23 | 0.35 | 0.30 | 0.35 | 0.22 (0.08) | 0.20 | 0.22 | 0.2 0 | 0 24 | 0.28 (0.12) | 0.35 | 0.28 | 0.32 | 0.17 |
| 1 Month | 0.69 (0.37) | 0.77 | 0.58 | 0.86 | 0.62 | 0.80 (0.34) | 1.00 | 0.80 | 0.56 | 0.85 | 0.57 (0.40) | 1.04 | 0.39 | 0.33 | 0 51 | 0.61 (0.50) | 1.03 | 0.22 | 0.26 | 0.96 | 0.58 (0.59) | 0.84 | 0.41 | 0.56 | 0.51 | 0.65 (0.50) | 0.81 | 0 97 | 0.55 | 0.75 | |
| 3 Month | 0.63 (0.22) | 0.67 | 0.58 | 0.77 | 0. 52 | 0.63 (0.29) | 0.57 | 0.54 | 0.68 | 0.74 | 0.48 (0.27) | 0.60 | 0.42 | 0.48 | 0.42 | 0.56 (0.43) | 0.92 | 0.30 | 0.20 | 0 86 | 0.55 (0.25) | 0.59 | 0.40 | 0.53 | 0.69 | 0.72 (0.49) | 0.71 | 1.03 | 0.33 | 0.81 | |
| 12 Month | 0.81 (0.18) | 0.76 | 0.90 | 0.76 | 0.80 | 0.83 (0.38) | 0.66 | 0.73 | 1.25 | 0.63 | 0.69 (0.29) | 0.94 | 0.50 | 0.75 | 0.56 | 0.62 (0.31) | 0.78 | 0.52 | 0.41 | 0.77 | 0.72 (0.27) | 0.84 | 0.47 | 0.76 | 0.82 | 0.73 (0.38) | 0.36 | 0.99 | 0.75 | 0.83 | |
| Australia | Overnight | … | … | … | … | … | … | … | … | … | … | … | … | … | … | … | … | … | … | … | … | 0.42 (0.32) | 0.44 | 0.30 | 0.25 | 0.71 | 0.57 (0.56) | 0.60 | 0.77 | 0.68 | 0.21 |
| 1 Month | 0.62 (0.32) | 0.68 | 0.36 | 0.81 | 0.47 | 0.70 (0.12) | 0.33 | 0.83 | 0.81 | 0.85 | 0.91 (0.67) | 1.32 | 0.76 | 0.98 | 0.62 | 0.85 (0.37) | 0.90 | 0.49 | 1.05 | 0.94 | 0.94 (0.42) | 1.15 | 0.88 | 0.91 | 0.83 | 1.36 (0.56) | 1.57 | 1.38 | 1.21 | 1.27 | |
| 3 Month | 0.74 (0.19) | 0.86 | 0.73 | 0.75 | 0.66 | 0.68 (0.40) | 0.18 | 0.54 | 0.97 | 0.74 | 1.05 (0.68) | 1.74 | 0.76 | 0.99 | 0.77 | 0.86 (0.30) | 0.87 | 0.60 | 1.03 | 0.93 | 0.78 (0.34) | 0.94 | 0.60 | 0.60 | 0.99 | 1.07 (0.41) | 1.01 | 1.16 | 0.99 | 0.85 | |
| Canada | 1 Month | 0.93 (0.36) | 0.88 | 1.00 | 0.91 | 0.90 | 0.99 (0.35) | 1.29 | 1.06 | 0.71 | 0.90 | 0.88 (0.44) | 0.53 | 0.63 | 1.33 | 1.02 | 0.96 (0.62) | 1.38 | 0.56 | 0.68 | 1.28 | 0.97 (0.49) | 0.93 | 1.28 | 0.90 | 0.78 | 1.19 (0.70) | 1.56 | 1.73 | 0.68 | 0.79 |
| 3 Month | 1.03 (0.42) | 0.70 | 1.06 | 0.99 | 1.22 | 0.93 (0.31) | 1.16 | 0.89 | 0.78 | 0.90 | 0.91 (0.30) | 0.73 | 0.69 | 1.22 | 0.99 | 0.94 (0.19) | 1.32 | 0.68 | 0.74 | 1.09 | 0.99 (0.36) | 0.70 | 1.25 | 1.14 | 0.79 | 0.94 (0.42) | 1.11 | 1.17 | 0.98 | 0.46 | |
| 12 months | 0.99 (0.32) | 0.69 | 1.12 | 0.84 | 1.14 | 0.72 (0.31) | 0.67 | 0.66 | 0.79 | 0.78 | 0.84 (0.35) | 1.02 | 0.77 | 1.11 | 0.44 | 0.94 (0.47) | 0.99 | 0.85 | 0.81 | 1.11 | 0.98 (0 27) | 1.03 | 1.13 | 0.92 | 0 85 | 1.01 (0.33) | 0.96 | 0.88 | 1.11 | 1.08 | |
| Singapore | Overnight | 0.67 (0.25) | 0.87 | 0.65 | 0.61 | 0.47 | 0.50 (0.20) | 0.57 | 0.46 | 0.54 | 0.44 | 0.67 (0.25) | 0.89 | 0.44 | 0.61 | 0.77 | 0.72 (0.15) | 0.89 | 0.88 | 0.65 | 0.50 | 0.31 (0.13) | 0.37 | 0.22 | 0.27 | 0.38 | 0.38 (0.20) | 0.54 | 0.45 | 0.30 | 0.23 |
| 1 Month | 0.83 (0.31) | 1.11 | 0.66 | 0.88 | 0.83 | 0.92 (0.45) | 0.78 | 0.88 | 1.17 | 0.82 | 0.79 (0.37) | 1.02 | 0.67 | 0.68 | 0.82 | 0.89 (041) | 0.76 | 1.05 | 0.88 | 0.87 | 0.63 (0 22) | 0.71 | 0.77 | 0.52 | 0.52 | 0.87 (0.48) | 1.28 | 1.02 | 0.78 | 0.39 | |
| 3 Month | 0.76 (0.30) | 0.91 | 0.63 | 0.85 | 0 74 | 0.90 (0.37) | 0.79 | 0.88 | 1.28 | 0.64 | 0.73 (0.32) | 0 .97 | 0.60 | 0.71 | 0.68 | 0.96 (0.36) | 0.78 | 1.15 | 0.89 | 1.02 | 0.59 (0.19) | 0.61 | 0.55 | 0.52 | 0.68 | 0.76 (0.40) | 1.01 | 0.97 | 0.58 | 0.42 | |
| 12 Month | 0.65 (0.34) | 0.86 | 0 72 | 0 59 | 0.55 | 0.82 (0.42) | 0.78 | 0 08 | 1 01 | 0 80 | 0.61 (0.29) | 0 95 | 0 56 | 0 56 | 0 39 | 0.78 (0 29) | 0 66 | 0 89 | 0.84 | 0.74 | 0.63 (0 29) | 0.57 | 0.75 | 0.09 | 0 53 | 0.57 (0 22) | 0.61 | 0.69 | 0 48 | 0.47 | |
| Korea | Overnight | 0.90 (0.36) | 0.81 | 0.99 | 1.08 | 0 67 | 0.97 (0.41) | 0 76 | 0 84 | 1.24 | 1 06 | 0.63 (0.27) | 0.65 | 0.71 | 0 60 | 0 72 | 0.56 (0.31) | 0 69 | 0.81 | 0 46 | 0.31 | 0.37 (0 24) | 0.53 | 0 41 | 0.14 | 0.20 | 0.33 (0 16) | 0 34 | 0.25 | 0 21 | 0 13 |
| Poland | Overnight | … | … | … | … | … | … | … | … | … | … | 0.54 (0.31) | 0.48 | 0.75 | 0.51 | 0.42 | 0.54 (0.34) | 0.58 | 0.44 | 0.72 | 0 42 | 0.71 (0.11) | 0 78 | 0.85 | 0.51 | 0.71 | 0.51 (0.28) | 0.41 | 0.50 | 0.63 | 0. 52 |
| 1 Month | … | … | … | … | … | … | … | … | … | … | 1.02 (0.49) | 0.85 | 1.25 | 1.21 | 0.71 | 1.17 (0.63) | 0.95 | 1 02 | 1.04 | 1.68 | 1.00 (0.57) | 1.35 | 0.96 | 0.73 | 0.99 | 1.09 (0.71) | 1.54 | 0.69 | 1.07 | 1.00 | |
| 3 Month | … | … | … | … | … | … | … | … | … | … | 1.27 (0.66) | 0 99 | 1.72 | 1.57 | 0.74 | 0.89 (0.60) | 0.95 | 0.79 | 0.93 | 0.91 | 0.81 (0.42) | 0.77 | 0..63 | 0.78 | 1.05 | 0.91 (0.18) | 0.91 | 0.97 | 0.89 | 0.83 | |
Market Efficiency Coefficients: Money Markets
| Maturity | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | |||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Ql | Q2 | Q3 | 04 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | ||
| U.S | Overnight | 0.30 (0.34) | 0.50 | 0.25 | 0.30 | 0.25 | 0.29 (0.18) | 0.37 | 0 24 | 0.26 | 0.31 | 0.29 (0.15) | 0.29 | 0.20 | 0.34 | 0.31 | 0.31 (0.11) | 0.23 | 0.35 | 0.30 | 0.35 | 0.22 (0.08) | 0.20 | 0.22 | 0.2 0 | 0 24 | 0.28 (0.12) | 0.35 | 0.28 | 0.32 | 0.17 |
| 1 Month | 0.69 (0.37) | 0.77 | 0.58 | 0.86 | 0.62 | 0.80 (0.34) | 1.00 | 0.80 | 0.56 | 0.85 | 0.57 (0.40) | 1.04 | 0.39 | 0.33 | 0 51 | 0.61 (0.50) | 1.03 | 0.22 | 0.26 | 0.96 | 0.58 (0.59) | 0.84 | 0.41 | 0.56 | 0.51 | 0.65 (0.50) | 0.81 | 0 97 | 0.55 | 0.75 | |
| 3 Month | 0.63 (0.22) | 0.67 | 0.58 | 0.77 | 0. 52 | 0.63 (0.29) | 0.57 | 0.54 | 0.68 | 0.74 | 0.48 (0.27) | 0.60 | 0.42 | 0.48 | 0.42 | 0.56 (0.43) | 0.92 | 0.30 | 0.20 | 0 86 | 0.55 (0.25) | 0.59 | 0.40 | 0.53 | 0.69 | 0.72 (0.49) | 0.71 | 1.03 | 0.33 | 0.81 | |
| 12 Month | 0.81 (0.18) | 0.76 | 0.90 | 0.76 | 0.80 | 0.83 (0.38) | 0.66 | 0.73 | 1.25 | 0.63 | 0.69 (0.29) | 0.94 | 0.50 | 0.75 | 0.56 | 0.62 (0.31) | 0.78 | 0.52 | 0.41 | 0.77 | 0.72 (0.27) | 0.84 | 0.47 | 0.76 | 0.82 | 0.73 (0.38) | 0.36 | 0.99 | 0.75 | 0.83 | |
| Australia | Overnight | … | … | … | … | … | … | … | … | … | … | … | … | … | … | … | … | … | … | … | … | 0.42 (0.32) | 0.44 | 0.30 | 0.25 | 0.71 | 0.57 (0.56) | 0.60 | 0.77 | 0.68 | 0.21 |
| 1 Month | 0.62 (0.32) | 0.68 | 0.36 | 0.81 | 0.47 | 0.70 (0.12) | 0.33 | 0.83 | 0.81 | 0.85 | 0.91 (0.67) | 1.32 | 0.76 | 0.98 | 0.62 | 0.85 (0.37) | 0.90 | 0.49 | 1.05 | 0.94 | 0.94 (0.42) | 1.15 | 0.88 | 0.91 | 0.83 | 1.36 (0.56) | 1.57 | 1.38 | 1.21 | 1.27 | |
| 3 Month | 0.74 (0.19) | 0.86 | 0.73 | 0.75 | 0.66 | 0.68 (0.40) | 0.18 | 0.54 | 0.97 | 0.74 | 1.05 (0.68) | 1.74 | 0.76 | 0.99 | 0.77 | 0.86 (0.30) | 0.87 | 0.60 | 1.03 | 0.93 | 0.78 (0.34) | 0.94 | 0.60 | 0.60 | 0.99 | 1.07 (0.41) | 1.01 | 1.16 | 0.99 | 0.85 | |
| Canada | 1 Month | 0.93 (0.36) | 0.88 | 1.00 | 0.91 | 0.90 | 0.99 (0.35) | 1.29 | 1.06 | 0.71 | 0.90 | 0.88 (0.44) | 0.53 | 0.63 | 1.33 | 1.02 | 0.96 (0.62) | 1.38 | 0.56 | 0.68 | 1.28 | 0.97 (0.49) | 0.93 | 1.28 | 0.90 | 0.78 | 1.19 (0.70) | 1.56 | 1.73 | 0.68 | 0.79 |
| 3 Month | 1.03 (0.42) | 0.70 | 1.06 | 0.99 | 1.22 | 0.93 (0.31) | 1.16 | 0.89 | 0.78 | 0.90 | 0.91 (0.30) | 0.73 | 0.69 | 1.22 | 0.99 | 0.94 (0.19) | 1.32 | 0.68 | 0.74 | 1.09 | 0.99 (0.36) | 0.70 | 1.25 | 1.14 | 0.79 | 0.94 (0.42) | 1.11 | 1.17 | 0.98 | 0.46 | |
| 12 months | 0.99 (0.32) | 0.69 | 1.12 | 0.84 | 1.14 | 0.72 (0.31) | 0.67 | 0.66 | 0.79 | 0.78 | 0.84 (0.35) | 1.02 | 0.77 | 1.11 | 0.44 | 0.94 (0.47) | 0.99 | 0.85 | 0.81 | 1.11 | 0.98 (0 27) | 1.03 | 1.13 | 0.92 | 0 85 | 1.01 (0.33) | 0.96 | 0.88 | 1.11 | 1.08 | |
| Singapore | Overnight | 0.67 (0.25) | 0.87 | 0.65 | 0.61 | 0.47 | 0.50 (0.20) | 0.57 | 0.46 | 0.54 | 0.44 | 0.67 (0.25) | 0.89 | 0.44 | 0.61 | 0.77 | 0.72 (0.15) | 0.89 | 0.88 | 0.65 | 0.50 | 0.31 (0.13) | 0.37 | 0.22 | 0.27 | 0.38 | 0.38 (0.20) | 0.54 | 0.45 | 0.30 | 0.23 |
| 1 Month | 0.83 (0.31) | 1.11 | 0.66 | 0.88 | 0.83 | 0.92 (0.45) | 0.78 | 0.88 | 1.17 | 0.82 | 0.79 (0.37) | 1.02 | 0.67 | 0.68 | 0.82 | 0.89 (041) | 0.76 | 1.05 | 0.88 | 0.87 | 0.63 (0 22) | 0.71 | 0.77 | 0.52 | 0.52 | 0.87 (0.48) | 1.28 | 1.02 | 0.78 | 0.39 | |
| 3 Month | 0.76 (0.30) | 0.91 | 0.63 | 0.85 | 0 74 | 0.90 (0.37) | 0.79 | 0.88 | 1.28 | 0.64 | 0.73 (0.32) | 0 .97 | 0.60 | 0.71 | 0.68 | 0.96 (0.36) | 0.78 | 1.15 | 0.89 | 1.02 | 0.59 (0.19) | 0.61 | 0.55 | 0.52 | 0.68 | 0.76 (0.40) | 1.01 | 0.97 | 0.58 | 0.42 | |
| 12 Month | 0.65 (0.34) | 0.86 | 0 72 | 0 59 | 0.55 | 0.82 (0.42) | 0.78 | 0 08 | 1 01 | 0 80 | 0.61 (0.29) | 0 95 | 0 56 | 0 56 | 0 39 | 0.78 (0 29) | 0 66 | 0 89 | 0.84 | 0.74 | 0.63 (0 29) | 0.57 | 0.75 | 0.09 | 0 53 | 0.57 (0 22) | 0.61 | 0.69 | 0 48 | 0.47 | |
| Korea | Overnight | 0.90 (0.36) | 0.81 | 0.99 | 1.08 | 0 67 | 0.97 (0.41) | 0 76 | 0 84 | 1.24 | 1 06 | 0.63 (0.27) | 0.65 | 0.71 | 0 60 | 0 72 | 0.56 (0.31) | 0 69 | 0.81 | 0 46 | 0.31 | 0.37 (0 24) | 0.53 | 0 41 | 0.14 | 0.20 | 0.33 (0 16) | 0 34 | 0.25 | 0 21 | 0 13 |
| Poland | Overnight | … | … | … | … | … | … | … | … | … | … | 0.54 (0.31) | 0.48 | 0.75 | 0.51 | 0.42 | 0.54 (0.34) | 0.58 | 0.44 | 0.72 | 0 42 | 0.71 (0.11) | 0 78 | 0.85 | 0.51 | 0.71 | 0.51 (0.28) | 0.41 | 0.50 | 0.63 | 0. 52 |
| 1 Month | … | … | … | … | … | … | … | … | … | … | 1.02 (0.49) | 0.85 | 1.25 | 1.21 | 0.71 | 1.17 (0.63) | 0.95 | 1 02 | 1.04 | 1.68 | 1.00 (0.57) | 1.35 | 0.96 | 0.73 | 0.99 | 1.09 (0.71) | 1.54 | 0.69 | 1.07 | 1.00 | |
| 3 Month | … | … | … | … | … | … | … | … | … | … | 1.27 (0.66) | 0 99 | 1.72 | 1.57 | 0.74 | 0.89 (0.60) | 0.95 | 0.79 | 0.93 | 0.91 | 0.81 (0.42) | 0.77 | 0..63 | 0.78 | 1.05 | 0.91 (0.18) | 0.91 | 0.97 | 0.89 | 0.83 | |
volatile than the MEC for the Fed funds rate. Higher interest rate volatility is sometimes used as an indicator of illiquid markets, but in the case of money markets, it should also be seen in context with the way the central bank intervenes.29 Appendix III shows that the correlation among the different segments of the money market is far from obvious, although there appears to be some correlation using bid-ask spreads in the case of Singapore.
C. Bond Markets
The bond markets can be classified according to issuer, i.e., government securities, mortgage-backed bonds, and corporate bonds. The secondary market for government securities is generally perceived as being the most liquid of the various bond markets. Government securities often play a special role as collateral and benchmarks for pricing of other securities, and as safe haven because of limited credit risk and the fact that the outstanding amounts often are quite large. In recent years, some countries have concentrated their public debt on fewer maturities but larger issues of each maturity with a view to promote the liquidity—often in light of standardized derivatives—rather than tailor the securities to the preferences of specific investors. A few issuers—sometimes when creating benchmarks—occasionally guarantee to buy back at a discount, thus ensuring the securities remain liquid. Finally, a transparent and credible public debt policy is conducive to creating liquid public debt markets.
Turnover ratios vary significantly among countries. Inoue (1999) conducted in 1997 a survey among the G-10 countries, and found the turnover ratio varied from almost 34 in France to 2½ in the Netherlands. The different turnover ratios are, for instance, affected by prudential regulation that may limit the amount of securities truly being available for trading; the extent to which the central bank uses government securities to conduct open market operations; if intraday liquidity for the payment system is provided in form of repurchase agreements or in form of pledging; etc. In emerging markets, it is not unusual to have turnover ratios below one. Then, the number of trades per day becomes a useful indicator. In countries where dealers are required to report to the stock exchange or where securities are dematerialized and the central depository collects information on final ownership, turnover information is sometimes available, while it is much more difficult to attain such information in other countries, in part because dealers usually consider such information a business secret.
Bid-ask spreads for government securities are only available for individual securities and derivatives. In the United States, the spreads in derivative markets are often smaller than in the cash market according to Fleming (2001). They are occasionally used as a proxy for how liquid the market for government securities really is, in part because the spread gives an indication of the hedging costs. Newly issued “on-the-run” benchmark securities typically have a lower spread than off-the-run securities, which are less traded. Accordingly, spreads are more useful as indicators of liquidity for different segments rather than for the whole market.
In the United States, the spreads of inter-dealer brokers are rather modest. The bid-ask spread for treasury bills has a median of 0.5 basis points with a range of 0–2 basis points (Fleming and Sarkar, 1999). According to Table 6, the yield-spread often increases with the maturity of the security, in part reflecting the inventory costs. On the other hand, long-term interest rates may be more stable than short-term interest rates, reflecting long-term expectations, and there may be higher turnover in longer securities, which contribute to a lower spread. Furthermore, the spread is often measured on the interest rate instead of the price. The same change in basis points of the interest spread is much larger than a similar change in the spread of the price because of the larger duration of longer bonds.30 This may also help explain the fact that in Singapore the spreads are lower for longer securities (Table 7). The spreads in thinner emerging markets are typically larger.31 But it is important to acknowledge that the spread is only a proxy and does not include all the costs of a securities transaction and varies significantly across markets.
Selected Liquidity Indicators for Government Securities Markets in G-10 Countries1/
Excluding Italy.
For 6-year bonds.
For 20-year bonds,
For 22-year bonds.
The figure is the mid-point of a range.
Tick size is shown in ten-thousandths of the face value of 100 currency units of each country. In many cases the tick size is the same for customers and interdealers.
Some or all securities are quoted in yield terms with the tick size is converted.
The figures may include trading other than outright transactions.
D means dealer market and A means auction-agency market.
Auction-agency markets exists.
Selected Liquidity Indicators for Government Securities Markets in G-10 Countries1/
| Canada | Belgium | France | Germany | Japan | Netherlands | Sweden | Switzerland | U.K. | U.S.A. | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Spreads: | On | Off | On | Off | On | Off | On | Off | On | Off | On | Off | On | Off | On | Off | On | Off | On | Off | ||
| Fixed coupon, | 2 years | 2 | … | n.a. | n.a. | 4 | n.a. | 4 | n.a. | 5 | n.a. | … | n.a. | 4 | n.a. | … | n.a. | 3 | … | 1.6 | n.a. | |
| 5 years | 5 | 135/ | n.a. | n.a. | 5 | 6 | 4 | 4 | 92/ | 112/ | n.a | n.a. | 9 | n.a. | … | n.a. | 4 | 4 | 1.6 | 6.3 | ||
| 10 years | 5 | 165/ | 5 | 7 | 10 | 6 | 4 | 5 | 7 | 7 | n.a | n.a. | 15 | n.a. | 10 | 10 | 4 | 4 | 3.1 | 6.3 | ||
| 30 years | 10 | 195/ | n.a. | n.a. | 24 | 10 | 10 | 10 | 163/ | 193/ | n.a | n.a. | 274/ | n.a. | 25 | n.a. | 8 | 12 | 3.1 | 12.5 | ||
| Index-linked, | 10 years | 25 | … | … | n.a. | … | … | … | … | … | … | … | … | 39 | n.a. | … | … | 15 | … | 6.3 | n.a. | |
| Tick size6/ | 0.1 | 1 | 1-27/ | 1 | 3-47/ | 1 | 2-117/ | 1 | 3.1 | 0.8-3.1 | ||||||||||||
| Turnover indicators | ||||||||||||||||||||||
| Turnover rate (a/b) | 21.9 | 4.1 | 33.8 | n.a. | 6.9 | 2.6 | 32.7 | 3.6 | 7.0 | 22.0 | ||||||||||||
| Yearly trading (U.S. million) (a) | 6,243 | 947 | 18,6348/ | n.a. | 13,282 | 4508/ | 3,626 | 1258/ | 3,222 | 75,901 | ||||||||||||
| Outstanding (U.S. million) (b) | 285 | 232 | 551 | 563 | 1,919 | 176 | 111 | 35 | 458 | 3,457 | ||||||||||||
| Outstanding volume, percent of GDP | ||||||||||||||||||||||
| Average issue size (U.S. billion) | ||||||||||||||||||||||
| Yearly trading volume in futures market (c) | 185 | 28 | n.a. | n.a. | 18,453 | … | 1137 | 90 | 3,294 | 27,928 | ||||||||||||
| Cash futures ratio (a)/(c) | 33.7 | 33.8 | n.a. | n.a. | 0.7 | … | 3.2 | 1.4 | 1.0 | 2.7 | ||||||||||||
| Market structure: | Custromer9/ | D | D10/ | D10/ | D110/ | D10 | A and D | D | A and D | D | D | |||||||||||
| Interdealer9/ | D | D | D | … | A and D | A | n.a. | … | D | D | ||||||||||||
| Memorandum items | ||||||||||||||||||||||
| Holdings of nonresidents (percent) | 25.0 | 23.0 | 12.9 | n.a. | 10.0 | 24.0 | 19.5 | n.a | 14.4 | 36.9 | ||||||||||||
Excluding Italy.
For 6-year bonds.
For 20-year bonds,
For 22-year bonds.
The figure is the mid-point of a range.
Tick size is shown in ten-thousandths of the face value of 100 currency units of each country. In many cases the tick size is the same for customers and interdealers.
Some or all securities are quoted in yield terms with the tick size is converted.
The figures may include trading other than outright transactions.
D means dealer market and A means auction-agency market.
Auction-agency markets exists.
Selected Liquidity Indicators for Government Securities Markets in G-10 Countries1/
| Canada | Belgium | France | Germany | Japan | Netherlands | Sweden | Switzerland | U.K. | U.S.A. | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Spreads: | On | Off | On | Off | On | Off | On | Off | On | Off | On | Off | On | Off | On | Off | On | Off | On | Off | ||
| Fixed coupon, | 2 years | 2 | … | n.a. | n.a. | 4 | n.a. | 4 | n.a. | 5 | n.a. | … | n.a. | 4 | n.a. | … | n.a. | 3 | … | 1.6 | n.a. | |
| 5 years | 5 | 135/ | n.a. | n.a. | 5 | 6 | 4 | 4 | 92/ | 112/ | n.a | n.a. | 9 | n.a. | … | n.a. | 4 | 4 | 1.6 | 6.3 | ||
| 10 years | 5 | 165/ | 5 | 7 | 10 | 6 | 4 | 5 | 7 | 7 | n.a | n.a. | 15 | n.a. | 10 | 10 | 4 | 4 | 3.1 | 6.3 | ||
| 30 years | 10 | 195/ | n.a. | n.a. | 24 | 10 | 10 | 10 | 163/ | 193/ | n.a | n.a. | 274/ | n.a. | 25 | n.a. | 8 | 12 | 3.1 | 12.5 | ||
| Index-linked, | 10 years | 25 | … | … | n.a. | … | … | … | … | … | … | … | … | 39 | n.a. | … | … | 15 | … | 6.3 | n.a. | |
| Tick size6/ | 0.1 | 1 | 1-27/ | 1 | 3-47/ | 1 | 2-117/ | 1 | 3.1 | 0.8-3.1 | ||||||||||||
| Turnover indicators | ||||||||||||||||||||||
| Turnover rate (a/b) | 21.9 | 4.1 | 33.8 | n.a. | 6.9 | 2.6 | 32.7 | 3.6 | 7.0 | 22.0 | ||||||||||||
| Yearly trading (U.S. million) (a) | 6,243 | 947 | 18,6348/ | n.a. | 13,282 | 4508/ | 3,626 | 1258/ | 3,222 | 75,901 | ||||||||||||
| Outstanding (U.S. million) (b) | 285 | 232 | 551 | 563 | 1,919 | 176 | 111 | 35 | 458 | 3,457 | ||||||||||||
| Outstanding volume, percent of GDP | ||||||||||||||||||||||
| Average issue size (U.S. billion) | ||||||||||||||||||||||
| Yearly trading volume in futures market (c) | 185 | 28 | n.a. | n.a. | 18,453 | … | 1137 | 90 | 3,294 | 27,928 | ||||||||||||
| Cash futures ratio (a)/(c) | 33.7 | 33.8 | n.a. | n.a. | 0.7 | … | 3.2 | 1.4 | 1.0 | 2.7 | ||||||||||||
| Market structure: | Custromer9/ | D | D10/ | D10/ | D110/ | D10 | A and D | D | A and D | D | D | |||||||||||
| Interdealer9/ | D | D | D | … | A and D | A | n.a. | … | D | D | ||||||||||||
| Memorandum items | ||||||||||||||||||||||
| Holdings of nonresidents (percent) | 25.0 | 23.0 | 12.9 | n.a. | 10.0 | 24.0 | 19.5 | n.a | 14.4 | 36.9 | ||||||||||||
Excluding Italy.
For 6-year bonds.
For 20-year bonds,
For 22-year bonds.
The figure is the mid-point of a range.
Tick size is shown in ten-thousandths of the face value of 100 currency units of each country. In many cases the tick size is the same for customers and interdealers.
Some or all securities are quoted in yield terms with the tick size is converted.
The figures may include trading other than outright transactions.
D means dealer market and A means auction-agency market.
Auction-agency markets exists.
Bid-Ask Spreads (Percent): Bond Markets
Bid-Ask Spreads (Percent): Bond Markets
| Country | Maturity | 1999 | 2000 | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | ||
| Singapore | 3 Month | 0.05% (0.01) | 0.05% | 0.05% | 0.05% | 0.05% | 0.06% (0.02) | 0.05% | 0.05% | 0.05% | 0.08% |
| 12 Month | 0.05% (0.00) | 0.05% | 0.05% | 0.05% | 0.05% | 0.05% (0.00) | 0.05% | 0.05% | 0.05% | 0.05% | |
| 5 Year | 0.01% (0.01) | 0.01% | 0.01% | 0.01% | 0.01% | 0.01% (0.01) | 0.01% | 0.01% | 0.01% | 0.02% | |
| 10 Year | 0.01% (0.01) | 0.01% | 0.01% | 0.01% | 0.00% | 0.01% (0.01) | 0.01% | 0.01% | 0.01% | 0.01% | |
| Malaysia | 3 Year | 0.22% (0.13) | 0.12% | 0.12% | 0.30% | 0.31% | 0.18% (0.11) | 0.29% | 0.20% | 0.15% | 0.12% |
| 5 Year | 0.21% (0.13) | 0.13% | 0.20% | 0.21% | 0.29% | 0.23% (0.10) | 0.30% | 0.23% | 0.21% | 0.19% | |
| 10 Year | 0.19% (0.23) | 0.08% | 0.10% | 0.32% | 0.24% | 0.26% (0.08) | 0.29% | 0.30% | 0.24% | 0.20% | |
Bid-Ask Spreads (Percent): Bond Markets
| Country | Maturity | 1999 | 2000 | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | ||
| Singapore | 3 Month | 0.05% (0.01) | 0.05% | 0.05% | 0.05% | 0.05% | 0.06% (0.02) | 0.05% | 0.05% | 0.05% | 0.08% |
| 12 Month | 0.05% (0.00) | 0.05% | 0.05% | 0.05% | 0.05% | 0.05% (0.00) | 0.05% | 0.05% | 0.05% | 0.05% | |
| 5 Year | 0.01% (0.01) | 0.01% | 0.01% | 0.01% | 0.01% | 0.01% (0.01) | 0.01% | 0.01% | 0.01% | 0.02% | |
| 10 Year | 0.01% (0.01) | 0.01% | 0.01% | 0.01% | 0.00% | 0.01% (0.01) | 0.01% | 0.01% | 0.01% | 0.01% | |
| Malaysia | 3 Year | 0.22% (0.13) | 0.12% | 0.12% | 0.30% | 0.31% | 0.18% (0.11) | 0.29% | 0.20% | 0.15% | 0.12% |
| 5 Year | 0.21% (0.13) | 0.13% | 0.20% | 0.21% | 0.29% | 0.23% (0.10) | 0.30% | 0.23% | 0.21% | 0.19% | |
| 10 Year | 0.19% (0.23) | 0.08% | 0.10% | 0.32% | 0.24% | 0.26% (0.08) | 0.29% | 0.30% | 0.24% | 0.20% | |
Table 8 shows the MEC for government securities with selected maturities in Australia, Canada, Singapore, and India during the period 1996-2000. There are significant differences between the different segments in each bond market and between countries, as also illustrated in the figures in Appendix IV.32 33
Market Efficiency Coefficients: Bond Markets
Market Efficiency Coefficients: Bond Markets
| Country | Maturity | 1997 | 1998 | 1999 | 2000 | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | ||
| Australia | 3 Month | 0.90 (0.62) | 1.42 | 0.86 | 0.89 | 0.59 | 0.62 (0.37) | 0.33 | 0.66 | 0.92 | 0.57 | 0.30 (0.19) | 0.48 | 0.31 | 0.24 | 0.50 | 1.02 (0.45) | 0.88 | 1.15 | 1.02 | 1.01 |
| 6 Month | 0.98 (0.60) | 1.50 | 0.89 | 0.82 | 0.75 | 0.63 (0.27) | 0.45 | 0.58 | 0.93 | 0.67 | 0.46 (0.26) | 0.48 | 0.28 | 0.28 | 0.77 | 1.06 (0.34) | 0.96 | 1.21 | 0.88 | 1.14 | |
| 1 Year | 0.68 (0.33) | 0.74 | 0.71 | 0.64 | 0.72 | 0.85 (0.51) | 0.70 | 0.78 | 1.28 | 0.83 | 0.38 (0.13) | 0.44 | 0.39 | 0.28 | 0.38 | 0.88 (0.43) | 0.73 | 0.61 | 0.62 | 0.69 | |
| 3 Year | 0.63 (0.24) | 0.58 | 0.55 | 0.62 | 0.77 | 0.91 (0.45) | 0.63 | 0.00 | 1.25 | 0.96 | 0.01 (0.28) | 0.76 | 0.91 | 0.80 | 2.14 | 0.87 (0.32) | 0.84 | 1.16 | 0.78 | 0.71 | |
| 5 Year | 0.81 (0.20) | 0.52 | 0.47 | 0.72 | 0.73 | 0.86 (0.26) | 0.66 | 8.87 | 1.01 | 0.87 | 0.85 (0.22) | 0.70 | 0.96 | 0.75 | 0.99 | 0.80 (0.28) | 0.62 | 1.09 | 0.02 | 0.67 | |
| 10 Year | 0.73 (0.28) | 0.66 | 0.52 | 0.79 | 0.72 | 0.83 (0.30) | 0.78 | 0.86 | 0.79 | 0.85 | 0.74 (0.17) | 0.89 | 0.68 | 0.84 | 0.75 | 0.75 (0.20) | 0.07 | 0.89 | 0.71 | 0.74 | |
| Canada | 3 Month | 0.80 (0.29) | 0.81 | 0.85 | 0.90 | 1.03 | 1.15 (0.50) | 1.48 | 0.84 | 1.01 | 1.29 | 0.98 (0.34) | 0.95 | 1.12 | 0.96 | 0.80 | 0.95 (0.40) | 1.03 | 1.00 | 0.97 | 0.97 |
| 6 Month | 0.93 (0.25) | 1.01 | 0.83 | 0.99 | 0.87 | 1.07 (0.47) | 1.40 | 0.90 | 0.68 | 1.11 | 1.00 (0.31) | 1.21 | 1.07 | 0.66 | 0.83 | 0.88 (0.34) | 0.98 | 1.02 | 0.70 | 0.87 | |
| 1 Year | 0.87 (0.31) | 1.07 | 0.83 | 0.70 | 0.87 | 1.02 (0.33) | 1.32 | 0.87 | 0.31 | 1.10 | 0.92 (0.29) | 1.12 | 0.93 | 0.86 | 0.79 | 1.03 (0.28) | 0.88 | 1.03 | 1.1.1 | 1.04 | |
| 3 Year | 0.88 (0.27) | 1.11 | 0.80 | 0.84 | 0.82 | 1.04 (0.32) | 1.24 | 0.88 | 0.90 | 1.17 | 0.95 (0.32) | 1.19 | 0.81 | 0.86 | 0.97 | 1.04 (0.32) | 0.64 | 1.00 | 1.00 | 1.24 | |
| 5 Year | 0.96 (0.25) | 1.23 | 0.80 | 0.90 | 0.90 | 1.12 (0.37) | 1.24 | 0.93 | 0.93 | 1.37 | 1.02 (0.34) | 1.28 | 0.84 | 1.00 | 0.99 | 0.97 (0.31) | 0.76 | 1.03 | 0.98 | 1.13 | |
| 10 Year | 1.01 (0.29) | 1.95 | 0.90 | 0.81 | 1.01 | 1.11 (0.40) | 1.23 | 0.97 | 0.94 | 1.32 | 1.03 (0.32) | l.22 | 0.81 | 1.14 | 0.94 | 0.92 (0.27) | 0.82 | 0.97 | 0.87 | 1.00 | |
| Singapore | 3 Month | 0.84 (0.32) | 0.77 | 1.11 | 1.14 | 0.71 | 1.07 (0.36) | 1.25 | 0.85 | 1.03 | 1.07 | 1.01 (0.51) | 1.14 | 0.57 | 1.13 | 1.22 | 0.88 (0.55) | 1.41 | 1.08 | 0 77 | 0.81 |
| 12 Month | 0.96 (1.34) | 0.58 | 0.58 | 1.58 | 0.68 | 0.93 (0.31) | 0.90 | 0.85 | 1.05 | 0.89 | 0.72 (0.45) | 0.79 | 0.38 | 1.05 | 0.88 | 1.07 (0.82) | 1.37 | 1.43 | 0.88 | 0.81 | |
| 5 Year | … | … | … | … | … | 1 02 (0.43) | 0 92 | 0.88 | 1.27 | 0.98 | 1.07 (0.58) | 1.49 | 0.70 | 0.93 | 1.13 | 0.78 (0.41) | 0.93 | 0 93 | 0.58 | 0.72 | |
| 10 Year | … | … | … | … | … | … | … | … | … | 0.69 | 0.99 (0.40) | 1.11 | 0.92 | 0.94 | 0.99 | 0.99 (0.37) | 0.77 | 1.06 | 1.15 | 1.01 | |
| India | 2 Year | 0.73 (0.60) | 0.39 | 1.60 | 0.47 | 0.51 | 0.50 (0.33) | 0.53 | 0.46 | 0.32 | 0.67 | 0.87 (0.53) | 0.97 | 0.72 | 0.98 | 0.79 | 0.98 (0.39) | 0.63 | 1.20 | 0.99 | 1.05 |
| 5 Year | 0.73 (0.54) | 0.48 | 1.20 | 0.52 | 0.69 | 0.60 (0.41) | 1.01 | 0.81 | 0.52 | 0.19 | 0.94 (0.50) | 0.55 | 0.86 | 1.40 | 0.89 | 0.84 (0.17) | 0.68 | 1.23 | 0.83 | 0.63 | |
Market Efficiency Coefficients: Bond Markets
| Country | Maturity | 1997 | 1998 | 1999 | 2000 | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | Avg | Q1 | Q2 | Q3 | Q4 | ||
| Australia | 3 Month | 0.90 (0.62) | 1.42 | 0.86 | 0.89 | 0.59 | 0.62 (0.37) | 0.33 | 0.66 | 0.92 | 0.57 | 0.30 (0.19) | 0.48 | 0.31 | 0.24 | 0.50 | 1.02 (0.45) | 0.88 | 1.15 | 1.02 | 1.01 |
| 6 Month | 0.98 (0.60) | 1.50 | 0.89 | 0.82 | 0.75 | 0.63 (0.27) | 0.45 | 0.58 | 0.93 | 0.67 | 0.46 (0.26) | 0.48 | 0.28 | 0.28 | 0.77 | 1.06 (0.34) | 0.96 | 1.21 | 0.88 | 1.14 | |
| 1 Year | 0.68 (0.33) | 0.74 | 0.71 | 0.64 | 0.72 | 0.85 (0.51) | 0.70 | 0.78 | 1.28 | 0.83 | 0.38 (0.13) | 0.44 | 0.39 | 0.28 | 0.38 | 0.88 (0.43) | 0.73 | 0.61 | 0.62 | 0.69 | |
| 3 Year | 0.63 (0.24) | 0.58 | 0.55 | 0.62 | 0.77 | 0.91 (0.45) | 0.63 | 0.00 | 1.25 | 0.96 | 0.01 (0.28) | 0.76 | 0.91 | 0.80 | 2.14 | 0.87 (0.32) | 0.84 | 1.16 | 0.78 | 0.71 | |
| 5 Year | 0.81 (0.20) | 0.52 | 0.47 | 0.72 | 0.73 | 0.86 (0.26) | 0.66 | 8.87 | 1.01 | 0.87 | 0.85 (0.22) | 0.70 | 0.96 | 0.75 | 0.99 | 0.80 (0.28) | 0.62 | 1.09 | 0.02 | 0.67 | |
| 10 Year | 0.73 (0.28) | 0.66 | 0.52 | 0.79 | 0.72 | 0.83 (0.30) | 0.78 | 0.86 | 0.79 | 0.85 | 0.74 (0.17) | 0.89 | 0.68 | 0.84 | 0.75 | 0.75 (0.20) | 0.07 | 0.89 | 0.71 | 0.74 | |
| Canada | 3 Month | 0.80 (0.29) | 0.81 | 0.85 | 0.90 | 1.03 | 1.15 (0.50) | 1.48 | 0.84 | 1.01 | 1.29 | 0.98 (0.34) | 0.95 | 1.12 | 0.96 | 0.80 | 0.95 (0.40) | 1.03 | 1.00 | 0.97 | 0.97 |
| 6 Month | 0.93 (0.25) | 1.01 | 0.83 | 0.99 | 0.87 | 1.07 (0.47) | 1.40 | 0.90 | 0.68 | 1.11 | 1.00 (0.31) | 1.21 | 1.07 | 0.66 | 0.83 | 0.88 (0.34) | 0.98 | 1.02 | 0.70 | 0.87 | |
| 1 Year | 0.87 (0.31) | 1.07 | 0.83 | 0.70 | 0.87 | 1.02 (0.33) | 1.32 | 0.87 | 0.31 | 1.10 | 0.92 (0.29) | 1.12 | 0.93 | 0.86 | 0.79 | 1.03 (0.28) | 0.88 | 1.03 | 1.1.1 | 1.04 | |
| 3 Year | 0.88 (0.27) | 1.11 | 0.80 | 0.84 | 0.82 | 1.04 (0.32) | 1.24 | 0.88 | 0.90 | 1.17 | 0.95 (0.32) | 1.19 | 0.81 | 0.86 | 0.97 | 1.04 (0.32) | 0.64 | 1.00 | 1.00 | 1.24 | |
| 5 Year | 0.96 (0.25) | 1.23 | 0.80 | 0.90 | 0.90 | 1.12 (0.37) | 1.24 | 0.93 | |||||||||||||