As regard to transaction taxes and volatility, the Finnish stock market has not been investigated very thoroughly. Joakim Westerholm has conducted a research, in which he investigated transaction costs’s impact on turnover, prices and volatility during the same observation period as this study. He compared the effects of FTT removal in Finland and Sweden. (Westerholm, 2003)
Westerholm used a dynamic asset-pricing model and his findings are in line with most of the empiric evidence, that a decrease in transaction costs leads to a decrease in volatility. He also found that lower transaction costs tend to rise the level of market activity. (Westerholm, 2003)
Finland had previously a transaction tax of 1 % that was levied from exchange trading and a slightly higher, 1,6% tax rate, which was levied on over the counter trades (Finlex, 2011). The tax was removed from exchange trading from the beginning of May 1992 and was invalidated in 1996. The law that abolished the tax from exchange trading was passed in the parliament on 30th of April, which did not leave very much time for the markets to react to the announcement. (Finlex, 2011)
When analysing the findings from Finland one must take into account, that at the time the tax was lifted, Finland was effectively in a recession.
3.1 The economic atmosphere around the tax reform
During the late 1980’s Finland enjoyed a time of strong economic growth and the economy was showing signs of overheating due to financial liberation and times of robust credit expansion. This era is commonly referred in Finland as the Casino Economy, which depicts the positive attitudes towards risk taking at the time.
The enthusiastic times of the late 1980 came to a grinding halt in the very early 1990 when the Finnish economy dived into a deep recession, caused by the simultaneous
house market and bank crises at a time when almost all of Finland’s exports to it’s biggest export country, Russia, were abolished with the demise of the Soviet union.
This kind of economic atmosphere can be difficult for analysing transaction costs effect on volatility, especially if Schwert’s findings about the historical volatility can be generalised to other markets (Schwert, W. 1990). If there is naturally significantly more volatility in times of economic turmoil and no way of separating fundamental volatility, this is a challenging era for conducting this kind of research. This is because it is the transitional part of volatility that should be affected by transaction costs, according to the reasoning of most academics.
3.2 Data and methodology
In order to investigate, the effects of the Finnish tax reform, one should have information about trade volume, daily index-value points and at least the relevant news count. However, no such data for this time period was available from the Nasdaq Helsinki Stock exchange, for this study.
Instead, data provided by the Bank of Finland was used. This data set comprises of 2510 observations of the Helsinki Stock Exchange OMX-index from 2nd of January 1987 to 31st December 1996. The index-values are closing values of the exchange and from them the daily revenue is calculated as the percentage of growth or decline.
The daily revenues were normally distributed.
The OMX-index contains all the stocks traded in the Helsinki Stock Exchange and they are weighted by their relative proportion in the index. Volatilities for 7 days, 15 days, one month, 30 days, one year and five years before and after the 1992 reform were calculated.
Figure one shows the development of the OMX-index with the respective daily returns over time from the observation period. The primary y-axis depicts the index
points on the left side of the chart and on the opposite side, the secondary y-axis shows the daily returns from the observation period.
Figure 1. OMX-index & daily returns
Figure two depicts the OMX-index from the same period as figure one, but instead of daily returns the secondary axis shows the seven-day volatilities from the observation period.
Figure 2. OMX-index and the seven-day volatilities
-‐8 %
Figure three shows the OMX-index together with the return volatility around the time of the reform giving a better picture about the possible effects of the change. The time of the reform is highlighted by a green oval on the graph.
Figure 3. OMX-index and the seven-day volatilities in 1992
From figure three we can see, that the FTT removal in Finland was followed by a moderate increase in the OMX-index, regardless of the general downward trend. This is in line with the findings of Umlauf. There is also a clear decrease in the volatility following the reform, which in line with the majority of empirical evidence in existence.
This decrease though, is followed by several months of highly volatile market conditions.
From figure 2., it is clear, that volatility tends to peak at the end of the year in the Helsinki Stock exchange, almost every year and the end of 1992 was especially volatile. Werterholm hints that the period of relatively high volatility at the end of 1992 would have more to do with the overall economic conditions than the FTT reform (Westerholm, 2003).
Table two in the other hand shows the differences in the stock exchange return volatility before and after the reform. This table summons the seven-day, 15-day, 30-day, one-month, one-year and five-year volatilities before and after the tax reform,
which was brought to bear 1st of May, 1992. The reason why there is a 30-day volatility calculated on top of the one month volatility is that the number of trading days in each month tends to vary, mainly because of holidays. The differences in volatility before and after the reform are shown on the last row. Short-term differences are highlighted with green and the long-term differences with red.
Table 2.
Time 5 yr Vola. 1 yr Vola. 1 month vol. 30 day vola. 15 day vola. 7 day vola.
Before 0,00914374 0,009969791 0,0132071 0,011751544 0,011041333 0,007265074
After 0,012571558 0,01469998 0,007163268 0,007894061 0,007926365 0,006202871
Diff. 0,003427818 0,004730189 -0,006043832 -0,003857483 -0,003114968 -0,001062203
From the table it is clear, that the five-year period preceding the tax reform has been somewhat less volatile than the period after the reform. This is also clearly visible from figure two. However, the short-term volatilities seem to be significantly higher before the reform for all time spans from seven days to thirty-day periods. This suggests that the removal of the Finnish transaction tax did lower volatility at least in the short run. This very interesting notion is in line with most of the empiric evidence.
Like noted above, it is clearly evident from figure three, that the later part of the year 1992 was increasingly volatile. If this were because of the change in the trader composition, that could be expected to show with a slight delay, the result would contradict with the arguments made by Lanne and Vesala (Lanne & Vesala, 2006).
One must now remember that these figures represent the whole market volatility during these periods, because the fundamental and transitional volatilities have not been separated from each other. This is also the case with many other studies around this field.