Skip to main content
SpringerLink
Log in

Securities Transaction Tax and Market Efficiency: Evidence from the Japanese Experience

  • Published:
Journal of Financial Services Research Aims and scope Submit manuscript

Abstract

This study examines the effect of transaction costs on the time series behavior of stock returns over a period surrounding the April 1989 changes in tax rates on securities transactions and capital gains in Japan. We find significant decreases in estimates of the first-order autocorrelation in returns for Japanese stocks listed in Japan, but no changes for Japanese stocks dually listed in the United States as American Depository Receipts (ADRs), which were not subject to the tax law change. We also find lower price basis between the ADRs and their underlying Japanese stocks. These results are consistent with the hypothesis that a reduction in transaction costs improves the efficiency of the price discovery process.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Figure 1
Figure 2
Figure 3
Figure 4

Similar content being viewed by others

Notes

  1. An anonymous referee graciously suggested this rationale.

  2. Another type of transaction cost comes in the form of bid–ask spread.

  3. From “Is there any reason to keep transaction tax?” The Nikkei Weekly, March 28, 1992.

  4. A potentially relevant factor is capital–loss deductibility. However, tax–loss deductibility in Japan, which allows for capital loss carryover for 3 years, was introduced only in calendar year 2003 (Proposed Reform to the Taxation of Capital Gain of Stocks, Ministry of Finance, Japan, October 30, 2001). Apparently, it is not applicable to our sample period.

  5. The 1% flat transactions tax option for satisfying the capital gains tax was explained as equivalent to a 20% tax on a 5% capital gain. A quick check on the Bank of Japan website (Statistics section) reveals that, in fiscal year 1989, the official discount rate was 4.25 or 4.5%, depending on the collaterals pledged, and the prime lending rate stood at 5.75% for short-term loans and 6.5% for long-term loans. In view of these concurrent interest rates, it is quite conservative for the Japanese tax authorities to assume the 5% rate of capital gain in the tax code, considering the higher risks involved in equity investments.

  6. From “Brokers hail securities transaction tax cut,” Jiji Press Ticker Service, December 15, 1995.

  7. From Japanese Tax System (1998 edition) and “Government approves taxation system reform for FY ‘98,” Japan Economic Newswire, January 9, 1998.

  8. From Japanese Tax System (1999 edition); FY 2003 Tax Reform (Main Points); “Government to bring forward commission decontrol to April,” Japan Economic Newswire, October 31, 1998; and “LDP panel wants securities transaction tax abolition in Jan.–April,” Jiji Press Ticker Service, December 14, 1998.

  9. Effective on April 1, 1998, brokerage commission on trades over 50 million yen was deregulated, and corporate tax rate was lowered by 3%. Starting on April 1, 1999, corporate tax rate was reduced by 4.5%, and full deregulation of brokerage commission was scheduled to take effect, but delayed to October 1.

  10. Trading shares were based on transactions on the First Section of the TSE for 1975–1980, and since 1981 on those on the First and Second Section of the TSE, as well as the Osaka and Nagoya Stock Exchanges.

  11. See, among others, Ito and Iwaisako (1995) and Ito (2003) for discussions about the possible causes of the asset price bubble in Japan.

  12. However, in most of the years, the estimated effective rates are somewhat higher than the official rates for some conceivable reasons. One clear reason is that the aggregate trading value excluded transaction values for foreign stocks listed and actively traded in Japan around the time, which were subject equally to the securities transaction tax. Another possible reason could be errors made in the bookkeeping, collection, aggregation, and reporting of the tax revenue and trading value. These factors may also help to account for the rate variations in the years (e.g., from 1982 through 1988) in between the official tax law changes.

  13. Indeed, using the same test in the corresponding sample periods for the tax reforms in 1978, 1981, and 1996, we find no significant efficiency impact in the Japanese equity market. Furthermore, for the three STT increases in 1973, 1978, and 1981, we do not have an adequate control sample, given that there were very few Japanese ADRs around the changes. For the STT increase in 1973, we do not even have the necessary data for the treatment (Japanese) stocks, given the common base date of 01/01/1973 for Japan in the Datastream and PACAP (Pacific-Basin Capital Markets), the only databases for Japan employed in the finance literature.

  14. Shrinking (extending) the sample period to 1 year (3 years) around the event day yields qualitatively similar results. Since the 1996 tax event is in between the two commission deregulations in April 1994 and April 1998, we tried only the shorter sample period to avoid potential contamination.

  15. Empirical results are qualitatively invariant to the weighting scheme (equal or value weighting). Therefore, to save space, we report only results from the equally weighted portfolios in the paper.

  16. The ADR returns are from the U.S. market, which lags the Japanese market. As such, we delay the event day (April 1, 1989) by one trading day, but find almost identical results, i.e., no efficiency changes occur to the ADRs around the tax reform. One could argue that some ADR volume might migrate to the home market following the event. This depends on if the transaction costs in Japan are lower than in the U.S. following the event. The lack of an efficiency effect on the ADRs does not seem to favor the migration argument.

  17. The U.S. does not levy taxes on security transactions at the federal level. However, as discussed in Belkin (2003), the State of New York imposed a tax on sales of corporate stocks and certificates in 1905, shifted the tax to New York City in 1966, added a surcharge in 1975, and began phasing out the tax in 1978. The effective elimination of the stock transfer tax was completed in October 1981, at which time 100% of the tax was rebated to the payer. We are grateful to an anonymous referee for guiding us to this information.

  18. Of course, cross-border arbitrage could indirectly affect the ADRs if the price discrepancies more than cover the arbitrage costs. Later, we show that the ADR portfolios are indeed unaffected, at least in terms of efficiency, by the tax reform in Japan.

  19. Qualitatively similar correlations also obtain in the out-of-sample periods (1 to 3 years in both directions) around the sample period.

  20. The only exception is the small-stock portfolio, TSESML, in the Pearson correlation.

  21. Following Neumark et al. (1991, Panel B, Table VI), we regress the home-market (TSEALL) returns on the ADR returns while controlling for the host-market returns, and find the following results: \( {\text{TSEALL}} = 0.0003 + 0.5624 * {\text{ADR}} - 0.1616 * {\text{S\& P}}500 \), with the two-sided t-statistics for the two regressors being respectively 24.965 and −5.873, both significant at the 1% level, and the adj-R 2 being 38.61%. The coefficient estimate for the ADR in this alternative equation is very close to that for the TSEALL in Eq. (1), suggesting a positive and significant correlation between the control and treatment portfolio return.

  22. To further check the robustness of our findings, we conduct the same tests using a hypothetical event day of April 1 in sample periods of similar lengths both before and after the real event day, April 1, 1989, but find no significant changes in first-order autocorrelation in either the control or the treatment portfolios. This exercise lends support to the efficiency effects of the tax reform.

  23. The daily price volatility is calculated as the intra-daily high–low price ratio in logarithm, i.e., ln(p h/p l), where p h and p l stand respectively for the intra-day high price and low price, as discussed in more detail in Parkinson (1980).

  24. The rate of increase in raw trading value was about 26% in terms of the daily trading value per firm in the two post-event years (1,147 million yen) over that in the two prior years (911 million yen), without adjusting for the concurrent inflations rates.

  25. The rate of decrease in the STT revenue was about 49.11% in terms of the average total STT revenue in the 2 years after the tax reform (990.5 billion yen) over that in the 2 years before the event (1,946.5 billion yen), even without taking into account the contemporary inflation rates.

  26. An anonymous referee kindly suggested to us this supplemental investigation.

  27. Using stocks in the TSEALL portfolio (covering all TSE stocks with continuous data in the sample period) and its three size-based sub-portfolios in the same 4-year sample period confirms the positive price effect of the tax event in the Japanese equity market. For instance, the average price for stocks in the TSEALL increased by about 20% (from ¥1,452 in the two prior years to ¥1,741 in the two post-event years).

  28. The switching regression model is equivalent to the mean-adjusted event study method, an arguably more appropriate one than the market-adjusted methods when investigating a marketwide event such as the tax changes in question. Brown and Warner (1980) explore these conventional event-study methods.

  29. We also try to estimate the price basis effect as the cross-sectional average of the changes in the firm-specific price basis around the event, and find almost identical results. The choice of closing and opening prices in Eq. 4 is dictated by the fact that trading in Japan precedes that in the U.S. We also test with closing prices for the ADRs, but find no material difference in results.

References

  • Anonymous, Is there any reason to keep transaction tax? The Nikkei Weekly, p. 14, March 28, 1992

  • Anonymous, Brokers hail securities transaction tax cut, Jiji Press Ticker Service, December 15, 1995

  • Anonymous, Government approves taxation system reform for FY ’98, Japan Economic Newswire, January 9, 1998

  • Anonymous, Government to bring forward commission decontrol to April, Japan Economic Newswire, October 31, 1998

  • Anonymous, LDP panel wants securities transaction tax abolished in Jan.–April, Jiji Press Ticker Service, December 14, 1998

  • Barberis N, Thaler R (2001) A survey of behavioral finance, Working paper, University of Chicago

  • Belkin D (2003) Revising the New York stock transfer tax: revenues and risks, Background Paper, New York City Independent Budget Office

  • Brown S, Warner J (1980) Measuring security price performance. J Financ Econ 8:205–258

    Article  Google Scholar 

  • Chou R, Wang G (2006) Transaction tax and market quality of the Taiwan stock index Futures. J Futures Mark 26:1195–1216

    Article  Google Scholar 

  • Eun C, Sabherwal S (2003) Cross-border listings and price discovery: evidence from U.S.-listed Canadian stocks. J Finance 58:549–575

    Article  Google Scholar 

  • Fama E (1970) Efficient capital markets: a review of theory and empirical work. J Finance 25:383–417

    Article  Google Scholar 

  • Fama E (1991) Efficient capital markets: II. J Finance 46:1575–1617

    Article  Google Scholar 

  • Froot K, Perold A (1995) New trading practices and short-run market efficiency. J Futures Mark 15:731–765

    Article  Google Scholar 

  • Grossman S, Stiglitz J (1980) On the impossibility of informationally efficient markets. Am Econ Rev 70:393–408

    Google Scholar 

  • Hu S (1998) The effects of the stock transaction tax on the stock market—Experiences from the Asian markets. Pac Basin Finance J 6:347–364

    Article  Google Scholar 

  • Ito T (2003) Retrospective on the bubble period and its relationship to developments in the 1990s. World Econ 26:283–300

    Article  Google Scholar 

  • Ito T, Iwaisako T (1995) Explaining asset bubbles in Japan, NBER Working Paper No. W5358

  • Jensen M (1978) Some anomalous evidence regarding market efficiency. J Financ Econ 6:95–101

    Article  Google Scholar 

  • Kupiec P (1995) A securities transactions tax and capital market efficiency. Contemp Econ Policy 13:101–112

    Article  Google Scholar 

  • Kupiec P (1996) Noise traders, excess volatility, and a securities transaction tax. J Financ Serv Res 10:115–129

    Article  Google Scholar 

  • Lo A, MacKinlay C (1988) Stock market prices do not follow random walks: evidence from a simple specification test. Rev Financ Stud 1:41–66

    Article  Google Scholar 

  • Lo A, Mamaysky H, Wang J (2004) Asset prices and trading volume under fixed transactions costs. J Polit Econ 112:1054–1091

    Article  Google Scholar 

  • Ministry of Finance, Japan, Japanese Tax System [1998 Edition] and [1999 Edition]

  • Ministry of Finance, Japan, Proposed Reform to the Taxation of Capital Gain of Stocks (Main Points), October 30, 2001

  • Ministry of Finance, Japan, FY 2003 Tax Reform (Main Points), December 19, 2002

  • Neumark D, Tinsley P, Tosini S (1991) After-hours stock prices and post-crash Hangovers. J Finance 46:159–178

    Article  Google Scholar 

  • Parkinson M (1980) The extreme value method for estimating the variance of the rate of return. J Bus 53:61–66

    Article  Google Scholar 

  • Tokyo Stock Exchange (1989) Fact Book

  • Tokyo Stock Exchange (1990) Fact Book

  • Tokyo Stock Exchange (2004) Fact Book

  • Umlauf S (1993) Transaction taxes and the behavior of the Swedish stock market. J Financ Econ 33:227–240

    Article  Google Scholar 

Download references

Acknowledgments

We are grateful to an anonymous referee and Haluk Ünal (the editor) for many constructive comments and suggestions, which have led to substantial improvements in the paper. We also wish to thank Youichirou Sone at the Japanese Ministry of Finance and Yuichi Kato at the Japanese National Tax Agency for kindly providing the securities transaction tax revenue data. Nonetheless, as usual, we are solely responsible for any errors or omissions remaining.

Author information

Authors and Affiliations

  1. Department of Economics and Finance, Texas A&M International University, Laredo, TX, 78041-1900, USA

    Shinhua Liu

Authors
  1. Shinhua Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Shinhua Liu.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Liu, S. Securities Transaction Tax and Market Efficiency: Evidence from the Japanese Experience. J Finan Serv Res 32, 161–176 (2007). https://doi.org/10.1007/s10693-007-0018-z

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10693-007-0018-z

Keywords

JEL Classifications

Search

Navigation