The Efficiency in Liquidity Measures during the US Monetary Announcements

We examine the impact of US economic news releases in the liquidity of eleven not so extensively researched emerging stock markets. We employ ten liquidity measures. The sample begins from June 2007 up to December 2016. Analysis is performed in a weekly frequency. China is the least liquid Asian market. Peru is the most liquid Latin American market. Most of the emerging markets are positively affected by the US news, offering diversification bene-fits to international investors. India and Argentina (China and Chile) are the Asian and Latin American countries with the highest (lowest) impacts, re-spectively. There is not a single best-in-class liquidity measure. The country with the lowest liquidity has the lowest impact from the US news releases. This result holds for both groups of countries in Asia and Latin America.

Theoretical Economics Letters investigating how they react on macroeconomic release announcements.
Evidence suggests that effects absorb fifteen minutes after the announcements and persist more in instruments with more liquidity. Fleming and Remolona [5] extended on how the prices and liquidity affected macroeconomic news releases.
Due to the strengths and weaknesses of each liquidity measure and proxy, we employ the ten most influential liquidity estimators to determine the efficiency and efficacy of each measure in representing emerging market liquidity. These are: turnover volume, two naive liquidity ratios, a conventional liquidity ratio, the Martin [6] liquidity index, Hui and Heubel [7] liquidity ratio, the Hasbrouck and Schwartz [8] market efficiency coefficient, the Hui and Heubel [7] Market-Adjusted liquidity, the Amihud [9] illiquidity index, the Liu [10] measure and the modified Amihud measure (Kang and Zhang, [11]).
There is overall informational efficiency, where the role of country-level institutional environments in the relation between information and stock prices is important (Dang, Moshirian and Zhang, [12]). Nikkinen et al. [13] was among the first and relatively recent papers to investigate such international efficiency in emerging stock markets through the effects of U.S. economic news releases on the market risk (volatility). The effect of U.S. economic news releases on the liquidity of Asian and Latin American emerging stock markets is answered in the present paper with an updated and different dataset as well.
This paper examines the efficiency of the ten most important liquidity measures in estimating liquidity of Asian and Latin American emerging stock markets, and then the time-series behavior for each liquidity measure of the eleven emerging stock markets. Finally, the impact of the most important US economic news releases affecting the liquidity of emerging stock markets is examined. The rest of the paper is organized as follows. Section 2 describes the data. Section 3 deploys the methodology. Section 4 describes the empirical results. Section 5 provides concluding remarks.

Data
Both Asian and Latin American markets have an increasing interest in literature, nowadays. Kim, Kim and Lee [14] found strong spillover effects in the Asian emerging markets occasionally causing US dollar liquidity problems. Kearney [15] and Agudelo, Giraldo and Villarraga [16]  Daily prices of all stock indices are employed and are also expressed in US dollar. Liquidity is measured in a weekly frequency, however. The impact of the US economic news is also accessed in a weekly frequency. Using no daily data to examine the impact of economic news releases are among others, Hardouvelis [17], Boyd et al. [18], Lamont et al. [19], Vortelinos and Gkillas (Gillas) [20] and Vortelinos and Gkillas [21].
is the change in weekly actual (realized) macroeconomic variable j (from week to week), d is the day of the week of the meeting, and D is the total number of days in that week. The variable Index is the observed tone of statement for any macroeconomic variable.  The US announcements concern the FED target rate as decided and communicated by the Board of governors of the Federal Reserve System. Usually, these announcements take place more than once a month regularly. The total number of the FED Funds Target Rate announcements employed by the present study, is twenty. This study employs only the most influential announcements within this period. The announcement time is at 10:00 am local time. We have not selected all FED rate announcements due to only few of the weekly FED announcements were not expected in the international financial markets.

Liquidity Measurement
This paper employs ten liquidity measures. All liquidity measures are estimated in a weekly frequency. Due to the small number of the observation in a weekly frequency we apply a non-parametric bootstrap approach in each liquidity measure, re-sampling from the original dataset (see Efron and Tibshirani, [25]).
The first liquidity measure is the turnover volume TV t L in t weeks; second measure is the turnover rate ( TR t L ): where TR t L is the turnover rate liquidity measure; TV t L is the turnover volume; t MC is the market capitalization which equals to close price times number of shares outstanding ( t t P S ⋅ ); and, third measure the liquidity ratio ( LR t L ): where t P ∆ is the close price changes; and TV t L is the turnover volume.
Another liquidity measure employed here, is the conventional liquidity ratio ( CLR t L ) as examined in Gabrielsen, Marzo and Zagaglia [26]. This measure provides a measure for how much traded volume is necessary to induce a price change of one percent. Volumes and prices are the key ingredients. The analytical expression of the liquidity ratio index is: where , i t S and , i t P are the number of trades and closing price of day i in week t. The liquidity ratio is usually computed for a number of assets and is aggregated over a pool with similar characteristics. The time interval (T, t) adopted to compute the index is typically chosen arbitrarily; in this paper, T equals to 5 (days) and t measures weeks. This means that large volumes of trades have little influence on price, for high values of conventional liquidity ratio ( CLR t L ). Obviously, this conceptual framework focuses more on the price aspect than on the issue of time or on the execution costs typically present in a market. Another liquidity measure is the Hui and Heubel [7] liquidity ratio ( HH t L ) which relates the volumes of trade to their impact on prices, and thus also to resiliency.
The lower the HH t L , the higher the liquidity.
where 5,t h and 5,t l are the highest and lowest logarithmic daily prices over last 5 trading days; t indicates the number of days (5-trading-days periods); 20,t V is the total turnover volume traded last 5 days; 5,t S is the number of trades within a 5-days period; and 5,t P is the average closing price over a 5-day period. Unrefined measures of liquidity could be nothing more than some kind of weighted average reflecting the frequency with which new information hits an index as compared with another. Hasbrouck and Schwartz [8] proposed the market efficiency coefficient ( MEC t L ) to distinguish short-term from long-term price changes. The MEC t L exploits the fact that price movements are more continuous in liquid markets, even if information is affecting equilibrium prices. Thus, for a given permanent price change, the transitory changes to that price should be minimal in resilient markets. The long-period variance is approximated by the weekly realized range; and the short-period variance is approximated by the average daily range over a week.
( ) where t is the number of week; T is the number of short periods in each longer is the long-period (weekly) volatility (realized range), as proposed by Martens, and van Dijk [28]: where , i t h and , i t l are the within the i-th daily high and low logarithmic prices for each t week (5 trading days period); and MEC t L tends to be closer but slightly below one in more resilient markets. Hui and Heubel [7] suggested the Market-Adjusted liquidity ( MA t L ) for equities. This liquidity measure (as any liquidity measure in this paper) is estimated in a weekly frequency. Firstly, the beta ( β ) coefficient is estimated in a weekly frequency via the capital asset pricing model (CAPM); where, the five daily returns of an aggregate emerging stock market index 1 are regressed on the twenty corresponding daily returns of 1 The aggregated emerging stock market index is a weighted average of all eleven emerging stock indices, examined in the present paper. Weights are selected depending on the contribution of each index's volume turnover to the summation. , m t R is the daily market (aggregated) return; β is the weekly regression coefficient (systematic risk); , i t u is the regression residuals (or specific risk).

,
, , where , i t R and , i t V are the daily-price return and the daily turnover volume accordingly, in day i and week t. This illiquidity index provides only a rough measure of the price impact on liquidity. A recent paper that examined the illiq t L illiquidity index is Gabrielsen, Marzo and Zagaglia [26] and Karolyi, Lee and Van Dijk [29]. Liu [10] introduced the standardized turnover-adjusted number of zero daily volumes over the prior x weeks ( Liu t L ). Results are reported only for the 1-week prior Liu's measure as reported in Kang and Zhang [11]: where TV t L is the t-week turnover; Defl is the same across all emerging markets and set to be ( ) where N is the number of non-zero trading volume days within week t;

Impact of News Releases
This sub-section reveals the methodology of the response of US economic news releases on liquidity of twelve emerging stock markets. Regressions, similar to those used in Chulia, Martens and van Dijk, [30] and Gospodinov and Jamali [31], are employed. Regressing dummy variables on the first differences of liquidity is employed here to examine the effect of news on liquidity: where , j t D is the dummy variable for the j category of news, Moreover, Newey and West's [33] heteroskedasticity and autocorrelation consistent (HAC) standard errors are used to ensure valid inference.  The results of the impact of US announcements on liquidity series are reported in Table 2 and Table 3     Notes. Table 2 reports the costant (a) and slope (β) coefficients as well as the adj R 2 , regarding the impact of US economic news releases on Asian emerging stock markets. * and ** indicate significance in the 10% and 5% significance level, respectively. Table 3. Impact of news releases, Latin America.   Table 3 [10] measure are the measures with the most statistically significant and highest in-absolute-terms impacts. So, there is not a single measure better than the others. respectively. It should be mentioned that the country with the lowest liquidity has the lowest impact from the US news releases. For Asia, this country is China; and for Latin America, this country is Chile. Moreover, there is not a single liquidity measure providing more significant impacts than the others, for both Asian and Latin American countries. So, a new liquidity measure specifically designed for emerging markets is needed to be introduced. It should also be considered that only the US news releases factor was employed to explain emerging markets liquidity. This is why the significant alpha (a) coefficient that incorporates all other explanatory factors of liquidity (except for US news) was statistically significant and negative in most of countries and liquidity measures.