Expiry day effects and arbitrage

Document Sample
Expiry day effects and arbitrage Powered By Docstoc
					    Do offshore markets provide a beneficial relationship to the
    onshore markets? An examination of the MSCI Taiwan Index
                                           Futures



                                 Professor Alex Frino1
            University of Sydney, Sydney, NSW 2006, Australia and CMCRC

                                    Wong, Jin Boon2
            University of Sydney, Sydney, NSW 2006, Australia and CMCRC




                           Working Paper (Preliminary Results)
                                  University of Sydney
                                    (Do Not Quote)




1
  Professor Alex Frino is currently the Chief Executive Officer of CMCRC (Capital Markets
Cooperative Research Centre) and a Professor of Finance at the University of Sydney. He has a PhD
in Finance from the University of Sydney and an MPhil in Finance from Cambridge University. Alex
has published over 40 research papers, one book and won over 2 million dollars in competitive
research funding. He has held visiting appointments at the SFE and Credit Suisse First Boston, and
maintains strong links with industry.
2
  Wong, Jin Boon is a PhD candidate at the University of Sydney and is currently a holder of the
prestigious CMCRC scholarship for high achievers. He has also previously won the Faculty of
Economics and Business Scholarship (USYD) for PhD candidates and R.J. Chamber Scholarship for
Honors students.
                                      Abstract


This paper examines the relationship between the MSCI Taiwan Index futures
contracts (MSCI-TW) that are traded on the Singapore Exchange (SGX), the
component stocks listed on the Taiwan Stock Exchange (TSE) and the TAIFEX
futures contracts (TX-FUT). We look at how the existence of arbitrage opportunities
between (a) MSCI-TW and its component stocks and (b) MSCI-TW and TX-FUT
creates an entwined relationship between the exchanges that is mutually beneficial.
Regression analysis using the daily turnover of these securities (in notional US
dollars) from the 31st May, 2003 to 31st May, 2008 documents that an increase in the
trading activities of the MSCI-TW leads to a corresponding increase in the turnover of
its component stocks and TX-FUT. We find that on average over the 5 year sample
period, MSCI-TW contributes towards 5.04% and 12.05% of the total turnover of its
component stocks trading and TX-FUT respectively. This would suggest that despite
the competitiveness amongst the onshore and offshore markets, there is a mutually
beneficial relationship that exists between them via the exploitation of arbitrage
opportunities by traders.
1. Introduction

Since the introduction of the Morgan-Stanley Capital International Taiwan Index
Futures (MSCI-TW) that is traded in the Singapore Exchange (SGX) on the 9th
January, 1997 and the Taiwan Stock Exchange Capitalization Weighted Stock Index
futures contracts (TX-FUT) that is traded in the Taiwan Futures Exchange (TAIFEX)
from the 21st July, 1998, the analysis of the relationship between the onshore and
offshore markets has been the subject of many researches.


Most studies to date tend to focus solely on the competition between the MSCI-TW
and TX-FUT with price discovery and information transmission amongst the two
future contracts being the most frequent topics (see Roope and Zurbruegg, 2002;
Chou and Lee, 2002; Chen et al., 2002; and Hsieh, 2004). Other researches comparing
the trading volume, price volatility and components of intraday bid-ask spreads of the
two futures were done by Huang (2002, 2004a and 2004b) and Webb, Muthuswamy
and Segara (2007). The expiry day effects were examined by Chung and Hseu (2008)
more recently but no studies thus far have examined how despite the competition, a
beneficial relationship may exist between the cross-border exchanges through the
exploitation of arbitrage opportunities between these securities.


The link between arbitrage activities and the trading volume of dual listed stocks were
first posited by Board and Sutcliffe (1996). In their paper, they propose that the
existence of two or more futures contracts on an index can lead to an increase in the
trading activities of all futures as spread arbitrage bring the prices of these assets into
line. Arbitrage opportunities occur whenever there is price disequilibrium in a
commodity that is traded in two or more market places. Risk free arbitrage occurs
when a transaction enables the arbitrageur to incur no negative cash flow at any
probabilistic or temporal state and a positive cash flow in at least one state. Board and
Sutcliffe (1996) posit that it is possible to perform risk free arbitrage between (a) spot
and future markets3, and (b) futures to futures market even though the securities may
be denominated in different currencies by hedging the foreign exchange risks. While

3
  Whilst, numerous studies have examined the relationship between spot-futures trading by using the
relative deviations of the future prices from their theoretical prices (see Billingsley and Chance, 1988;
Brenner, Subrahmanyam and Uno 1989; and Butterworth and Holmes, 1999), they are based on the
onshore market with the same underlying currency. Board and Sutcliffe posit that spot-futures arbitrage
is possible even when the futures are denominated in different currencies.
they highlight that arbitrage activities that are exposed to currencies risks are less
attractive to traders, it is worthwhile to mention that the stock composition of the
TAIEX Index makes it more viable for arbitragers to exploit MSCI Taiwan spot-
futures arbitrage despite the currencies risks4.


This paper seeks to contribute to the literature by examining how the existence of
arbitrage opportunities between (a) MSCI-TW and its component stocks and (b)
MSCI-TW and TX-FUT creates an entwined relationship between the exchanges that
is mutually beneficial. Our empirical results show that an increase in the turnover of
MSCI-TW is associated with a corresponding increase in the trading activities of the
components stocks and TX-FUT. These findings provide a deeper insight into the role
of offshore markets and suggest that despite the competition the offshore market
provides, a beneficial relationship to the onshore markets exists in the form of
arbitrage trading by market participants. Relevant exchanges and regulators may find
it worthwhile to redevelop their business strategies and cooperate to promote trading
activities in Taiwanese related products via arbitrage exploitation.


The remainder of the paper is set out as follow. Section 2 provides the hypothesis
development. Section 3 provides details on the institutional setting. Section 4
describes the data and methodologies used. Section 5 presents the empirical results
and Section 6 provides a summary and conclusion on the findings.


2. Hypothesis Development

For arbitrage opportunities to exist between (a) MSCI-TW and its component stocks
and (b) MSCI-TW and TX-FUT, there must be a mean-reverting long-run equilibrium
relationship between the pairs. Prior literature analyzing the relationship between the
cross-border securities document that the returns of the two indices have highly
correlated return series which is above 0.975. More importantly, researchers analyzing
the price discovery between the MSCI-TW, TX-FUT and their component stocks like

4
  As of 31st May 2008, the TAIEX Index includes all of the 687 listed companies in TSEC and the
MSCI Taiwan Index comprises only 114 small, medium and large stocks (based on liquidity) in TSEC.
This smaller stock composition in the MSCI Taiwan Index makes it easier and more efficient for spot-
futures arbitrageurs to replicate and is hence more likely to be traded despite the currencies risks.
5
  Huang (2002) and Webb, Muthuswamy and Segara (2007) observed that the returns correlation
between the TAIEX Index and MSCI Taiwan Index were above 0.98 in their data sample and Hsieh,
2004 noted a similar relationship in excess of 0.97.
Roope and Zurbruegg (2002), Chou and Lee (2002), Chen et al. (2002), and Hsieh,
2004 observed the existence of cointegrated relationships6. This suggests the
existence of arbitrage opportunities and its exploitation7.


Arbitrage trading is beneficial to the cross-border exchanges as it requires the
simultaneous buying of a long position and selling of a short position. When traders
exploit any disequilibrium in the spot-futures pricing, they will buy MSCI-TW (or its
component stocks) and sell its component stocks (or MSCI-TW). This implies that
when the trading volume of MSCI-TAI (in SGX) increases due to arbitrage trading,
the trading activities of the component stocks in Taiwan Stock Exchange (TSE) will
also increase. This leads to our first hypothesis,


H1: An increase in the turnover of MSCI-TW will lead to an increase in the turnover
of its components stocks, ceteris paribus.


Similarly, arbitrage exploitation between the MSCI-TW and TX-FUT will also lead to
an increase in the trading activities of both futures, suggesting that,


H2: An increase in the turnover of MSCI-TW will lead to an increase in the turnover
of TX-FUT, ceteris paribus.


3. Institutional Setting

The Morgan-Stanley Capital International Taiwan Index Futures (MSCI Taiwan)
contracts that are traded in the Singapore Exchange (SGX) originated on the 9th
January, 1997. Initially, the index was based on the market capitalization index of 77
component stocks traded on the Taiwan Stock Exchange Corporation (TSEC) but this
have increased to 114 stocks as of 30th May, 2008. The Taiwan Stock Exchange
Capitalization Weighted Stock Index (TAIEX) futures contracts that are traded in the

6
  The concept of cointegration was used by Engle and Granger (1987) to describe the opportunity of
arbitrage. In situations where the prices of different assets are cointegrated, a mean reverting
relationship exists (e.g. whenever the prices deviate from their equilibrium relationship, arbitrage
behavior will bring the price relationship back to their equilibrium). That means non-stantionary
financial time series may possess long-run equilibrium relationships and have a tendency to move
together in the long-run by arbitrage activities.
7
  Consistent with the findings of previous research, we found that the returns correlation for the two
indices in our sample period is above 0.98 and both (a) and (b) were found to be cointegrated.
      Taiwan Futures Exchange (TAIFEX) were initiated on the 21st July, 1998. This index
      future comprises all of the listed shares on the Taiwan Stock Exchange (TSE).


      Table 1 below provides the contracts specifications of these two index futures.


           MSCI Taiwan Index Futures                                 Taiex Index Futures
        Items                 Specifications                   Items                 Specifications
                                                                                Taiwan Stock Exchange
Underlying Stock Index    MSCI Taiwan IndexSM            Underlying Index        Capitalization Weighted
                                                                                  Stock Index (TAIEX)
    Ticker Symbol                   TW                    Ticker Symbol                    TX

                         US$100 x MSCI Taiwan
    Contract Size                                          Contract Size            NT$200 x Index
                         IndexSM Futures price
    Minimum Price            0.1 index points             Minimum Price              One index point
      Fluctuation         (equivalent to US$10)             Fluctuation                (NT$200)
 Trading Hours (Local                                  Trading Hours (Local
                             08:45AM-1:50PM                                         08:45AM-1:45PM
         Time)                                                 Time)
                           Second last business                                 The third Wednesday of
  Last Trading Day          day of the contract          Last Trading Day        the delivery month of
                                  month.                                             each contract


                                                                                  The first business day
                                                                                following the last trading
                                                                                   day. All of the open
 Final Settlement Day        Last trading day.         Final Settlement Day      interests after the final
                                                                                 settlement day shall be
                                                                                    settled on the final
                                                                                     settlement price



                                                                                The final settlement price
                                                                                     for each contract is
                                                                                  computed from the first
                                                                                  fifteen-minute volume-
                                                                                    weighted average of
                         The final settlement price
                                                                                 each component stock's
                              shall be the official
                                                                                prices in the index on the
                         closing price of the index,
Final Settlement Price                                 Final Settlement Price    final settlement day. For
                           rounded to the nearest
                                                                                 those component stocks
                          two decimal places, on
                                                                                that are not traded during
                             the last trading day.
                                                                                   the beginning fifteen-
                                                                                   minute interval on the
                                                                                final settlement day, their
                                                                                 last closing prices would
                                                                                     be applied instead.


     Settlement              Cash settlement                Settlement              Cash settlement
                                                                                          Combined with MTX of
                                                                                            the contract-size
                                                                                                adjusted

                                                                                              1.Any investor's
                                                                                          aggregate open same-
                                                                                           side positions in the
                                                                                           Contract for various
                                                                                          delivery months at any
                        A person shall not own or                                       time shall not exceed the
                        control more than 10,000                                               limit standards
Position Limit          contracts net long or net              Position Limit               announced by the
                           short in all contract                                                   TAIFEX
                           months combined.                                               2.Institutional investors
                                                                                              may apply for an
                                                                                            exemption from the
                                                                                           above limit on trading
                                                                                           accounts for hedging
                                                                                                   purpose

                                                                                          3.These position limits
                                                                                           are not applicable to
                                                                                            omnibus accounts




  4. Data and Methodology

  The data used in the present study were supplied by the Securities Industry Research
  Centre of Asia-Pacific (SIRCA). The data includes the daily closing prices, volume
  traded and turnover of the TAIFEX futures contracts, the MSCI Taiwan Index futures
  contracts and its component stocks. The daily closing prices of the MSCI Taiwan
  Index were also obtained. The sample period starts from 31st, May 2003 to 31st May,
  2008.

  To determine the impact of trading activity (volume) in the MSCI-TW on the trading
  activity of its components stocks, we estimate the following equation using ordinary
  least-squares (OLS) regression:


  MSCI Taiwan Stockst = α + β1 MSCI-TWt + β2 Non-MSCI Taiwan Stockst + εt                               (1)


  where MSCI Taiwan Stockst is the combined turnover8 of the MSCI Taiwan Index
  component stocks in time period t; MSCI-TWt is the combined turnover of the MSCI-

  8
   Notional Value is computed as Price * Volume * Dollar Value * FX Rate, where price is the closing
  price; volume is the total volume traded for the day; dollar value is the dollar value of each index point
TW contracts in time period t; Non-MSCI Taiwan Stockt is the combined turnover of
the stocks listed on the TSE that are not part of the MSCI Taiwan Index in time period
t.


This regression analyses how a change in one US dollar turnover of the MSCI-TW
affects the turnover of its component stocks. Stocks listed on the TSE that are not part
of the MSCI Taiwan Index are included in the regression as controls variables to
account for the broad market movements of Taiwan related securities.


To examine how changes in the trading volume of the MSCI Taiwan futures affect the
trading activities of the TX-FUT, we estimate the following equation using ordinary
least-squares (OLS) regression:


                  TX-FUTt = α + β1 MSCI-TWt + β2 TSE Stockst + εt (2)


where TX-FUTt is the combined turnover of the TAIFEX future contracts in time
period t; MSCI-TWt is the combined turnover of the MSCI Taiwan futures contracts
in time period t; and TSE Stockt is the combined turnover of all stocks listed on the
TSE in time period t.

This regression analyses how a change in one US dollar turnover of the MSCI-TW
affects the turnover of TX-FUT. All stocks listed on the TSE are used as controls for
the general market movements in the Taiwan securities market.


From the results obtained in (1) and (2), the percentage contribution to the component
stocks of the MSCI Taiwan Index and TX-FUT is calculated as follows:


              CtMTSp = (β1(1) * MSCI-TWp) / MSCI Taiwan Stocksp                        (3)


where CtMTSp is the percentage contribution of MSCI-TW towards MSCI Taiwan
Index component stocks over the sample period p; β1(1) is the coefficient of MSCI-
TWp in equation (1); MSCI-TWp is the combined turnover of the MSCI-TW contracts


in local currency and fx rate is the daily exchange rate to the US dollars provided by the US Federal
Reserve.
   over the sample period p and; Non-MSCI Taiwan Stockp is the combined turnover of
   the stocks listed on the TSE that are not part of the MSCI Taiwan Index over the
   sample period p.


                   CtTFp = (β1(2) * MSCI-TWp) / TX-FUTp                      (4)


   where CtTFp is the percentage contribution of MSCI-TW towards TX-FUT over the
   sample period p; β1(2) is the coefficient of MSCI-TWp in equation (2); MSCI-TWp is
   the combined turnover of the MSCI-TW contracts over the sample period p and; TX-
   FUTp is the combined turnover of the TAIFEX future contracts over the sample
   period p.


   CtMTSp and CtTFp provide an approximation on the percentage contribution by
   MSCI-TW on its component stocks and TX-FUT, respectively, due to arbitrage
   trading.


   5. Results

                             Table 2: Regression Parameter Estimates
         Panel A: MSCI Taiwan Index Futures and MSCI Taiwan Index Component Stocks
                                           MSCI Taiwan       Non-MSCI Taiwan
                                                                                        Constant
                                             Futures              Stocks
MSCI Taiwan
                       Coefficients            0.09                 0.88               968,632,126
  Stocks
                      Standard Error           0.01                 0.03               36,066,627

                       T-statistics          6.24****            31.69****              26.86****


                         Panel B: MSCI Taiwan Index Futures and TX-FUT
                                           MSCI Taiwan
                                                               TAIEX Stocks             Constant
                                             Futures

  TX-FUT               Coefficients            0.17                 0.48                3,554,841

                      Standard Error           0.02                 0.02               55,119,574

                       T-statistics          9.15****            28.93****                0.06



              Panel C: Contribution of MSCI-TW to its Component Stock and TX-FUT (%)
                            MSCI Taiwan Index Component Stocks                    TX-FUT

      MSCI-TW                                    5.04%                             12.05%
**** denotes significance at the 0.001 levels.



         Panel A provides the results for equation (1). The results document a positive and
         statistically significant coefficient of 0.09, this shows that after controlling for the
         broad market movements of Taiwan related securities, an increase in the turnover of
         the MSCI-TW leads to an increase in turnover of the component stocks. Specifically,
         an increase of one US dollar in turnover of the MSCI-TW will lead to an approximate
         0.09 US dollar equivalent increase in the turnover of its component stocks.


         Panel B presents the results for equation (2). Similar to (1), the results documents a
         positive and significant coefficient of 0.17, this demonstrates that after controlling for
         general market movements in the Taiwan securities market, an increase of one US
         dollar in turnover of the MSCI-TW leads to a 0.17 US dollar equivalent increase in
         the turnover of TX-FUT. The percentage contribution by MSCI-TW is presented in
         Panel C. Results show that on average 5.04% of MSCI Taiwan component stocks’
         turnover and 12.05% of TX-FUT’s turnover can be attributed to MSCI-TW.


         In summary, these results show that a portion of the trading activities of MSCI-TW,
         its component stocks and TX-FUT are inter-linked through arbitrage trading and
         suggests that despite the competition between the three exchanges in these alternative
         investment products, a beneficial relationship exists amongst them.


         6. Conclusion

         This study analyzed the relationship between the (a) MSCI-TW and its component
         stocks and (b) MSCI-TW and TX-FUT for the 5 year period from 31st May, 2003 to
         31st May, 2008. Whilst numerous studies have examined the competitiveness amongst
         the exchanges in terms of price discovery and information share, we analyze how the
         existence of an additional offshore listed future contract benefits the trading activities
         of the onshore market. Empirical evidence shows that an increase in the turnover of
         MSCI-TW leads to an increase in the turnover of both its component stocks and TX-
FUT. We postulate that the link in the turnover of the futures and the component
stocks is due to arbitrage trading by market participants. In light of these results, it
may be of interest to the relevant regulators and exchanges to formulate a business
strategy that involves promoting arbitrage trading amongst the dual-listed securities
(and the component stocks) to entice global institutions to trade in Taiwan related
securities.


References

Board, J., Sutcliffe, C., 1996, The dual listing of stock index futures: Arbitrage,
spread arbitrage and currency risk, The Journal of Futures Markets, 16, 29–54.

Chen, S.-Y., Lin, C.-C., Chou, P.-H., Hwang, D.-Y., 2002, A comparison of hedge
effectiveness and price discovery between TAIFEX TAIEX index futures and SGX
MSCI Taiwan index futures, Review of Pacific Basin Financial Markets and Policies,
5, 277–300.

Chou, R.K., Lee, J.-H., 2002, The relative efficiencies of price execution between the
Singapore exchange and the Taiwan futures exchange, The Journal of Futures
Markets, 22, 173–196.

Chung, H., Hseu, M-M, 2008, Expiration day effects of Taiwan index futures: The
case of the Singapore and Taiwan Futures Exchanges, Journal of International
Financial Markets, Institutions and Money, 18, 107-120.

Hsieh, G.W.-L., 2004, Regulatory changes and information competition: The case of
Taiwan index futures, The Journal of Futures Markets, 24, 399–412.

Huang, Y.C. (2002), Trading activity in stock index futures markets: The evidence of
emerging markets, Journal of Futures Markets, 22, 983–1003.

Huang, Y.C. (2004a), The market microstructure and relative performance of Taiwan
Stock Index Futures: A comparison of the Singapore Exchange and the Taiwan
Futures Exchange, Journal of Financial Markets, 7, 335–350.

Huang, Y.C. (2004b), The components of bid-ask spread and their determinants:
TAIFEX versus SGX-DT, Journal of Futures Markets, 24, 835–860.

Roope, M., Zurbruegg, R., 2002, The intraday price discovery process between the
Singapore exchange and Taiwan futures exchange, The Journal of Futures Markets,
22, 219–240.
Webb, I., Muthuswamy. J., Segara, R., 2007, Market microstructure effects on
volatility at the TAIFEX, The Journal of Futures Markets, 27, 1219-1243.