1 Studentnumber: s1717383
Name: Joost Jonk Study Program: Msc Finance
Supervisor: Raymond Zaal
The influence of politics on the stock market: the
Dutch case
Abstract: Politics have an influence on the returns on the stock market. The theory behind the political business cycle effect states that incumbent policymakers can introduce new policies which will increase stock market returns and improve their chance on re-election. Another theory is the partisan effect. This theory states that returns on the stock market are either higher under left-wing or right-wing governments. Results in previous studies are mixed and this paper will contribute to this research field by testing political effects on the Dutch stock market. This research paper finds that there is a political business cycle effect in the Netherlands and that returns on the stock market are higher under a right-wing government compared to a left-wing government.
1.0 Introduction
In a democracy, policy decisions reflect the preferences of the majority of the voters. To ensure that the policy makers reflect the majority of the voters, the voters have to vote for these policy makers on a regular basis. When there is a change in the composition of votes this will result in a change in the composition of the government which in turn will result in a change in policies. These changes in policies can affect economic variables such as unemployment, growth and inflation and these changes in policies will have an effect on investor’s expectations, and thus on the returns on the stock market. (Vuchelen, 2003).
2 election results are known. However in a multi-party coalition government it is harder to predict future policy changes at the moment the election results are known since the composition of the government is still unknown. Election results may, however, still have some effects on the stock market since the uncertainty related to the event “election” is eliminated (Vuchelen, 2003). This paper will investigate whether Dutch elections and the policies of the composed government affect the performance of the Dutch stock market. Previous studies have already shown the impact of elections in the USA and found a ‘presidential business cycle’ where stock markets perform better in the last two years of the president’s term compared to the first two years (Huang, 1985; Johnson, Chittenden & Jensen, 1999). Other effects of elections on the stock market were found in studies about the difference in left-wing and right-wing governments, the so called ‘partisan effect’. Several authors found that stock markets perform better under a left-wing in the United States (Gärtner and Wellershoff, 1995, 1999; Santa-Clara and Valkonov, 2003) while other studies found that stock markets have a higher return under a right-wing government in New Zealand and the United Kingdom (Hudson, Keasey & Dempsey, 1998; Cahan, Malone, Powell & Choti, 2005). This paper will
contribute to this research by testing the effect of the political business cycle and the partisan effect for the Dutch stock market.
.
2.0 Literature Review
2.1 Stock market
The stock market index aggregates the individual stock prices traded on that index. Uncertainty puts a penalty on these individual stock prices, investors discount their expectations concerning possible future developments (Vuchelen, 2003). In an efficient market these individual stock prices fully reflect all available information (Fama, 1970, 1991). The stock prices are right in the sense that they reflect market participants’ rational assessments of business fundamentals (Fama, 1970; Summers, 1986). Under these assumptions, the price of a stock will equal its fundamental value. This
3 investor. Rational investors will consider all news available to them and government policies are obviously part of that news (Vuchelen, 2003).
Investors can either have a long-term or a short-term view. Short-term investors will only take the future expected cash flows in the short run into account. Policy changes will have an effect on the stock prices for this group of investors since it will change these future expected cash flows. For long-term view investors, a temporary change in policies only exerts a marginal effect on stock prices. For example when there is a policy introduced that reduces the expected future cash flows in the short run, but increase the expected future cash flow in the long run. This policy will increase the stock market index if the net present value of the expected cash flows in the long run will exceed the net present value of the expected cash flows in the short run (Vuchelen, 2003). However there is an uncertainty surrounding new policies, and since there are elections every four years in the Netherlands the new government may change some or all policies. Therefore it can be assumed that the short-term effects will frequently dominate the long-term effects (Vuchelen, 2003).
2.2 Does the stock market anticipate political events?
Ample research has been done about the impact of politics on the stock market. Most of this research has been done with US data. Some authors studied elections and found a four-year election cycle (Allvine & O’Neill, 1980; Huang, 1985; Johnson et al., 1999; Booth & Booth, 2003; Sturm, 2009). This four year election cycle implies that stock returns are higher during the second half of the governments’ term than during the first half (Sturm, 2013). Several authors argue that this is due to the fact that time remaining to election influences the administration’s economic policy decisions, which in turn influences stock prices. The idea behind this theory is that the incumbent government implements policies to stimulate the economy and the stock market in order to achieve a higher chance of re-election (Stovall, 1992; Booth & Booth, 2003; Zhao, Liano & Hardin, 2004).
Other studies find other results. For example Foerster (1994) finds effects on Canadian stock returns when there are changes in Canadian or American government changes. Foerster and Schmitz (1997) found that there is a transnational effect of US elections on the stock market index of 18 countries. Döpke and Pierdzioch (2006) tested the effect of the political business cycle on stock prices in Germany but could not find significant results. From these results it can be stated that political events can have an influence on the stock market.
4 used since the US has a party government. In countries with two-party systems and a single-party government all uncertainty is immediately removed when the election results are definitive and investors can anticipate on the policies this single-party will implement (Vuchelen, 2003).
However in countries with a proportional election system, like the Netherlands, governments are mostly multi-party coalitions. In these countries elections do not lead to a straightforward prediction of the composition of the new coalition since forming the coalition takes some time after the elections are final (Vuchelen, 2003). This implies that there is a variable time lag between elections and the stock market reactions since the stock market will react when the formation of the multi-party government is complete. Another point is that coalition-based political systems are less stable since government changes can occur without an election taking place, for example when one party in the coalition does not agree to a new policy but one or more opposition parties do. Then the composition of parties which accept a new policy is changed (Vuchelen, 2003).
2.3 Political business cycle effect
The theory behind the effect of the political business cycle is that the popularity of the government depends on the state of the economy (Döpke & Pierdzioch, 2006) and was first introduced by Down (1957) and Nordhaus (1975). Nordhaus (1975) states that voters have a certain expectation of the party which is in control. When this incumbent party reaches or exceeds this expectation then the voter will vote for this party again. When the incumbent party does not fulfill the voters expectations then the voter will vote for another party, the party which they think will fulfill their expectations, when the new election is there. The idea behind this theory is that voters consider their financial situation when voting. Voters have a preference for either left-wing parties which are preferring employment or right-wing parties which tend to favor low-inflation (Döpke & Pierdzioch, 2006; Herron, 2000; Hibbs, 1977). Each voter can align himself in a natural way with the party whose preferences most coincides with his own (Nordhaus, 1975). This leaves the ‘swing voters’ to control the election outcome. The problem with this approach is that the party which sticks the most to its principles, either low unemployment or low inflation, will lose to the party which is more adaptive since the more adaptive party will win the most swing votes. Since political parties are only interested in election outcomes and they know the voters preferences, they will choose economic policies during its incumbency which maximize its chance at re-election (Down, 1957; Nordhaus, 1975).
5 create a rising stock market by changing the policy instrument. When the stock market is rising this obviously will affect the financial situation of the voters. Therefore policymakers try to maximize their probability of re-election by following expansionary policies that generate low unemployment and high economic growth before the elections. After the elections the new government can
implement policies which raise taxes and cut government expenses which will lead to lower returns on the stock market. This is possible since the new government has the full four years until the new elections and they only have to stimulate the economy prior to the elections. (Köksal & Çalışkan, 2012).
This should lead to higher stock market returns in the last two years of the political business cycle compared to the first two years. Several authors have indeed found that the stock market return is higher during the last two years of the political business cycle compared to the first two years (Allvine & O’Neill, 1980; Huang, 1985; Stovall, 1992; Gärtner & Wellershoff, 1995; Johnson et al., 1999; Siegel, 2002; Booth & Booth, 2003; Jones & Banning, 2009; Sturm, 2009).
Based on this theory stating that the stock market returns are higher during the last two years of the political business cycle compared to the first two years, the following hypothesis is presented: H1: A political business cycle effect can be identified in the Netherlands.
2.4 Partisan effect
6 left-wing parties. If everything stays equal, an extension of a right-wing government will imply stable, if not rising, stock prices. An expected switch from a left-wing to a right-wing government could stimulate stock prices further since investors might assume that the new coalition will implement some supply-side policies (Vuchelen, 2003). However, empirical results of the partisan effect are mixed. Studies from Gärtner and Wellershoff (1995, 1999) and studies from Santa-Clara and
Valkonov (2003) show that when a left-wing party has the power there are higher stock returns in the US compared to when a right-wing party has the power. Döpke and Pierdzioch (2006) did not find a significant difference on German stock market returns between left-wing and right-wing government while Bohl and Gottschalk (2006) did found a better stock market performance under a left-wing government in Germany but also in Denmark and the US. Cahan et al. (2005) found different results with New Zealand data, they found that the stock market performs better with a right-wing
government compared to a left-wing government. Hudson et al. (1998) found that stock markets returns are higher in the UK when a right-wing government is in charge compared to a left-wing government.
From all these studies it remains unclear under which party (left or right) stock returns are higher. Since the results are mixed the researcher takes the position that returns are higher under a right-wing government compared to a left-wing government. This leads to the following hypothesis:
H2: Returns on the stock market are higher during a right-wing government compared to a left-wing government
2.5 Majority versus proportional representation and political events
There is a difference between majority-based electoral systems and proportional representation. The first results in a single-party government while the latter results in a coalition-based political system. This distinction needs to be made to analyze the impact of political events on the stock market. In a majority-based electoral system there are usually two parties. When there are only two parties the main political event are the elections. When elections are finished all uncertainty around the next government and their preferred policies is removed. Investors are then able to predict future policies fairly accurate on the basis of election results (Vuchelen, 2003).
7 the investors can get information about which party is the biggest, and more often than not this party will be in the government. Investors can already adjust their expectations on future policies with this information (Pantzalis, Stangeland & Turtle, 2000).
In coalition systems there are more political events compared to the two-party system. This is due the fact that not all uncertainty about future policies is removed with the election outcome. According to Vuchelen (2003) four political events can affect the stock market and investors adjust their
expectations after each event.
1. The election results. The first information the investor receives about future policies are the election results.
2. The time required to form a coalition. The formation of a new coalition requires an unknown amount of time.
3. The composition of the coalition. The relation between the election results and the
composition of the coalition is, by definition, not necessarily unambiguous. In most of the cases the new coalition will be either right-wing or left-wing.
4. The new government policies. This is of crucial importance for investors. The composition of the coalition already gives an indication about their policy priorities, but investors have to wait until the program of the coalition has been made public to really obtain a complete and correct view on future policies.
According to Vuchelen (2003) a change in investors’ expectations concerning one or more of the four events and their uncertainties will have an immediate effect on the stock market. For example when the formation of a new coalition takes a longer period than expected this can be a sign that the coalition will not be optimal. Therefore investors can become pessimistic, resulting in a negative effect on the stock market. Because there are four events, the election results itself will have a limited impact on the stock market since not all uncertainties are eliminated with the elections. The impact on the stock market will depend on the impact that these elections have on the investors’ expectations of the next coalition (Vuchelen, 2003). The election results will show that the next coalition will
8 2.6 Reaction of individual investors to policy expectation
In a system with proportional representation investors receive information on multiple moments. Investors will already form expectations of the policies of the new possible governments and will adjust their investments to that, which will be reflected in the stock market. There is however a lot of uncertainty which is partly removed when the election results are final. At that moment the investors know which party had the most votes and will be most likely in the government. It is not until the actual formation of the coalition that all uncertainty around future policies is removed. The formation of a coalition has taken an average of 72 days in the period 1945-2012 in the Netherlands1. This implies that there is a longer period of uncertainty in systems with proportional representation compared to systems with two parties.
Vuchelen (2003) gives an example on how this information is processed for an individual investor. Suppose an election is held at time t. In this election there are three participating parties which can form two possible coalitions, either right-wing or left-wing. These coalitions can come up with two kinds of policies, either R or L. As stated before the election result itself does not imply knowledge of the coalition. These are two separate events and must be separated. The parties require a time, T, to form a new coalition.
According to Vuchelen (2003), just before the election at time t-1, an investor will form an
expectation of the policies of the new government, either R or L. The investor expects the coalition to be formed at t+T. This expectation will be:
Where is the expected policy at time t-1 of the new coalition formed at time t+T; and are the expected policies at time t-1 in the case of a right- or left wing government formed at time t+T; and and are the probability (as perceived by the investor at time t-1) that a right- or a leftwing government will be formed at time t+T ( + = 1).
The expectation that the investor bases on these expected policies is part of the information he
possesses. This information will be reflected in the stock prices. A change in stock prices is caused by new information. According to several authors (Forsythe, Nelson, Neumann & Wright, 1992, 1995;
1
9 Jacobsen, Potters, Schram, van Winden & Wit, 2000; Shum, 1995) changes in the expected policies before the election is held, or in the variance of these expectations, are new information.
Investors will predict the result of the potential impact of the formation of the coalition before the election is held. The outcome uncertainty will disappear once the election results are final.
The stock prices however will respond in line with the expectations of the investor with regard to anticipated future policies. According to Pantzalis et al. (2000) and Vuchelen (2003) the effect of the election outcome on expected policy is:
When rearranged:
The change in the expected policy as a result of the outcome of the election reflects the change in the probabilities attached to the two potential coalitions ( and ) and the change in expected policies of these coalitions ( and ) (Vuchelen, 2003). The time that is required to form a new coalition and the uncertainty around this process is not included in this equation.
Elections in a coalition-based political system are less decisive compared to its two-party counterpart but these elections do remove some uncertainty with regards to future policies. This implies that a small effect on the stock market cannot be excluded. However when the election result implies a formation of previously impossible coalitions or to a political blockage, the stock market will probably react negatively (Vuchelen, 2003).
When the coalition is formed this will remove the uncertainty with regard to future policies in the same way as the election results in a two-party system. However there is still the question how long it takes to form the coalition whereas in a two-party system it is known with certainty when the
government formation will occur. According to Vuchelen (2003) the prediction error may be obtained by comparing the policy outcome to the predicted policy at the time of the election:
10 Which can be rearranged when using :
The idea behind this formula is that the prediction error equals the policy forecast errors weighted by the different probabilities. When a good analysis of the party programs has been made by an investor in advance then or will be small. This however does not imply that there will be no forecasting error since the probabilities are just as important, and these are harder to predict. According to Vuchelen (2003) a prediction error is quite likely in practice so the formation of a new coalition will always affect the stock market since the announcement of this coalition contains some degree of news. Vuchelen (2003) states that the reaction of the stock market will be immediate and proportional to the importance of the news.
3.0 Methodology
3.1 Political business cycle effect
To test for the existence of the political business cycle effect in the Netherlands, the research method of Döpke & Pierdzioch (2006) and MacCallum (1978) was followed. MacCallum (1978) developed a test to empirically analyze political business cycles in the US. It is relevant to also use the research method by Döpke and Pierdzioch(2006) since they used the MacCallum test in Germany. Germany has the same multi-party government type as the Netherlands where the US has not.
This test uses the returns on the stock market index of the AEX ( ) for the sample period. To test for the returns per quarter a dummy variable has been used. Every four year period (or less) a
government is in charge, is divided in quarters. Starting in the quarter in which an election is held this dummy variable assumes the values 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16 or 17. This implies that for each quarter between elections the dummy variable has a different value. This dummy variable can therefore be used to test whether systematic and statistically significant election cycles can be detected in stock market returns. Since there is an average formation period of 72 days, which is approximately 1 quarter, the government is in charge from the dummy which takes the value of 1. Table 1 summarizes which dummies represent the years the government is in charge.
Year 1 Year 2 Year 3 Year 4
11 In the sample period ten governments were in charge so every dummy variable has ten or less values for the regression. For example when there are early elections, because a government failed to complete the four years in office, there are less than 16 dummies for this period. There is one period where there were late elections and this period has 17 dummies.
Döpke & Pierdzioch (2006) state that this test is not directly applicable to German data since the political institutions in Germany, and also in the Netherlands, are different from the US. The election terms are not as regular as in the US and the length of an election term may be different due to unexpected elections. For this reason terms can be shorter or longer than the 16 proposed by
MacCallum (1978). In order to account for such variations the dummy variable has to be set equal to zero when an election occurs. Döpke & Pierdzioch (2006) propose the following regression model to test for the presence of an election cycle:
Where denotes stock market returns, denotes the McCallum dummy and denotes the error term. The McCallum dummy should be used since Political Business Cycle theory implies that an election cycle should be detectable at the level of a stock market index. Since the equation is formulated in terms of stock market returns, the first difference of the McCallum dummy should be used (Gärtner & Wellershoff 1995, 1999). It is also important to implement a crash test dummy to the right side of the equation, this dummy will cover periods of ‘stock market crashes’. This is necessary to not let extreme exogenous variables influence the regression result. Döpke & Pierdzioch (2006) count every quarter in which stock market returns exceed -20% as a ‘crash’ period and this research follows the same criterion.
After the regression is performed, the returns per quarter can be compared to see whether a political business cycle effect is detectable in the Netherlands. When the returns of the quarters in the last two years (dummies 9 to 17) are significantly higher than the returns of the quarters in the first two years (dummies 1 to 8), then there is a political business cycle effect present in the Netherlands.
3.2 Partisan effect
12 regression approach using quarterly data is the most appropriate methodology since in a coalition-based system the impact of several political events on the stock market should be tested. This is also in line with the average formation period of 72 days and the theory of multiple political events on which investors form their expectations. The formation period and the political events happen most of the time in the first quarter after the elections.
As stated before the research method of Vuchelen (2003) is followed, the quarterly percentage change in the stock price index of the AEX is explained by the percentage of change in the S&P index (as mentioned earlier the American stock market influences other stock markets in the world), the change in the business cycle index, the change in the bond rate and the change in the dollar exchange rate, which will be the control variables. Also political variables are used, L is a dummy variable equal to 1 when a left-wing party is the largest party in the coalition, R is a dummy variable equal to 1 when a right-wing party is the largest party in the coalition. ELEC is a dummy variable set to 1 in the quarter an election is held. The regression is:
13 4.0 Data
The research about the political business cycle contains a time span of 30 years between 1983 and 2013. The Netherlands has a coalition based political system and in the sample period 10 governments held office. 3 governments made the full 4 year period and 7 governments failed to make it to the end of the term. In the sample period 9 elections were held.
The sample data is smaller for the partisan effect since not all the data of the economic variables taken into account are from 1983 until 2013, instead they are from 1988 until 2013. In this sample period of 25 years 8 governments held office. In the sample period 7 elections were held. From these 8
governments that held office, 2 had a left party which was the largest and 6 had a right party that was the largest.
To test for the partisan effect a division had to be made between left and right governments. To decide whether a party is either left or right the database from Beck, Clarke, Groff & Walsh (2001) is used. Beck et al. (2001) made a political database on all countries in the world. Their database contains all political parties in charge of their countries and the political parties which are in opposition. This database covers the years from 1950 until now and the database shows which party is right, left or centre2. Beck et al. (2001) based the division of left, right and centre on a scientific way in their paper which supports the choices for their database. When following this database the result was that there were only centre-right or purple governments(coalition composed of both left and right parties). Since this division complicates straight forward application of the partisan theory an adjusted division has been applied. The division between a right and left government has been based on the party with the most power in the coalition of the government.
It is worth noting that a sudden great decrease in stock price due to exogenous events can influence the regression results. Therefore the crash dummy has been used in the political business cycle test and for the partisan effect the quarters with a return of -20% or less were not taken into account in the regression analysis.
Economic data (interest rates, index prices, exchange rates) has been retrieved by using Datastream. The economic business cycle indicator has been extracted from the site of ‘De Nederlandse Bank’ (DNB)1.DNB provides the business cycle indicator for the last 25 years on their website and after sending an e-mail to the DNB they provided an additional 5 years of data.
The software program Eviews has been used for the statistical analysis of the regression models.
2Source:
14 5.0 Results
5.1 Political Business Cycle
According to theory a political business cycle can be detected by McCallum’s (1978) dummy
approach. Table 2 shows the results of this test. The variable ‘crash’ represents the crash dummy and the variables ‘1’ to ‘17’ represents the quarters in dummies. The crash dummy is significant at the 1% level and negative. Dummies representing the 6th, 14th ,15th ,16th and 17th quarter are significant at the 5% level while dummies representing the 2nd, 10th and 11th quarter are significant at the 10% level. Since the dummy with the value of 0 has been left out to avoid multicollinearity, the coefficient of the other dummies should be interpreted as the difference in the coefficient compared to the coefficient of the dummy with the value of 0. This does not make a difference in interpretation since it is the difference in the coefficients of the dummies which are needed for the research. A higher coefficient means that dummy represents higher stock market returns in that quarter compared to the dummy with the value of 0.
15 Variable Coefficient Prob.
Constant -0,007130 0,7891 Crash -0,276093 0,0000 1 0,013449 0,7740 2 0,048408 0,0990 3 0,000832 0,9817 4 0,039094 0,2660 5 0,028126 0,5662 6 0,062639 0,0353 7 0,069363 0,1229 8 0,031348 0,4614 9 0,061514 0,1623 10 0,081319 0,0882 11 0,096566 0,0972 12 0,062305 0,1118 13 0,043606 0,2930 14 0,097096 0,0095 15 0,138887 0,0002 16 0,118182 0,0000 17 0,093666 0,0006 R-squared 0,446187 Adjusted R-squared 0,351247 Durbin-Watson stat. 1,962369
Table 2: Regression results for testing the political business cycle effect
5.2. Partisan Politics
16 compared to the regression without a lag. The model has an squared of 0.6963 and an adjusted R-squared of 0.6758 which gives the model explanatory power. The Durbin-Watson statistic of 2.001484 shows that the residuals are independent of one another. White and ARCH tests were performed to find heteroscedasticity and autocorrelation but no significant results were found. With regard to the partisan effect there is a difference between a right or left government, with the difference in dummy coefficients being 0.000591 in favor of the right government. Therefore hypothesis 2 is accepted, the stock market in the Netherlands performs better under a right-wing government in charge compared to a left-wing government.
Variable Coefficient Prob.
S&P 0,928941 0,0000 Cyclus -0,009416 0,0860 Interest rate 0,018024 0,0946 Exchange rate 0,362853 0,0001 ELEC(D)-1 -0,023845 0,1149 RIGHT(D) 0,025748 0,0302 LEFT(D) 0,025157 0,0770 R-squared 0,6963 Adjusted R-squared 0,6758 Durbin-Watson stat. 2,001484
Table 3: Regression results for testing the partisan effect
6.0 Conclusion
The goal of this research was to find statistical evidence of politics influencing the returns on the Dutch stock market. This research has indeed found some interesting results of the influence of politics on the stock market in the Netherlands. A political business cycle effect has been found, according to theory the returns should be higher in the last two year of the cycle compared to the first two year. This research found that in the first fifteen quarters every quarters return exceeded the return of the previous one. Only the returns in the 16th and 17th quarter are lower than their preceding
17 can be stated that politicians do influence the stock market by introducing policies which will increase their chance on re-election. It could be interesting for another research paper to test if these
policymakers indeed get re-elected when a political business cycle effect is present.
As for the partisan effect, stock market returns in the Netherlands tend to be higher under a right-wing government compared to a left-right-wing government. This is in line with results from New Zealand, the United Kingdom and Belgium (Hudson et al., 1998; Vuchelen, 2003; Cahan et al., 2005). This result is however not in line with results from the United States where the stock market returns are higher under a left-wing government (Gärtner and Wellershoff, 1995, 1999; Santa-Clara and Valkonov, 2003).
According to theory the stock market should perform better under a right-wing government compared to a left-wing government (Hibbs, 1977; Herron, 2000). It is strange that empirical findings are not always in line with theory. In some countries the stock market performs better under right-wing governments while in other countries the stock market performs better under left-wing governments. It is possible that there is another variable which is influencing the partisan effect, for example the political system. It is also possible that the current theory is not a good one and needs to be revised. It could be interesting for another research paper to find out what the influence of the political system is on the partisan effect.
18 References
Alesina, A., Sachs, J., 1988. Political parties and the business cycle in the United States, 1948 -1984. Journal of Money, Credit and Banking, Vol. 20 Issue 1, p. 63-82
Allvine, F., O’Neill, D., 1980. Stock market returns and the presidential election cycle, implications for market efficiency. Financial Analysts Journal, Vol. 36, No.5, p.49-56
Beck, T.G., Clarke, A., Groff, P.K., Walsh, P., 2001. New tools in comparative political economy: the database of political institutions. World Bank Economic Review, Vol. 15 Issue 1, p.165-176
Booth, J.R., Booth, L.C., 2003. Is presidential cycle in security returns merely a reflection of business conditions?. Review of Financial Economics, Vol.12, Issue 2, p131-159
Bohl, M.T., Gottschalk, K., 2006. International evidence on the democrat premium and the presidential cycle effect. The North American Journal of Economics and Finance, Vol. 17 Issue 2, p.107-120
Cahan, J., Malone, C.B., Powell, J.G., Choti, U.W., 2005. Stock market political cycles in a small, two-party democracy. Applied Economic Letters, Vol. 12 Issue 12, p.735-740
Döpke, J., Pierdzioch, C., 2006. Politics and the stock market: evidence from Germany. European Journal of Political Economy, Vol. 22 Issue 4, p.925-943
Down, A., 1957. An economic theory of democracy. Harper and Row, New York
Fama, E.F., 1970. Efficient capital markets: A review of theory and empirical work. Journal of Finance, Vol. 25 Issue 2, p.383-417
Fama, E.F., 1991. Efficient capital markets: II. Journal of Finance, Vol. 46 Issue 5, p.1575-1617
Foerster, S., 1994. Stock market performance and elections: made in Canada effects?. Canadian Investment Review, p.39-42
Foerster, S., Schmitz, J., 1997. The transmission of US election cycles to international stock returns. Journal of International Business Studies, Vol. 28 No. 1, p1-27
19 Forsythe, R., Frank, M., Krishnamurthy, V., Ross, T.W., 1995. Using market prices to predict
elections results: the 1993 UBC election stock market. Canadian Journal of Economics, Vol. 28 Issue 4, p.770-793
Gärtner, M., Wellershoff, K.W., 1995. Is there an election cycle in American stock returns? International Review of Economics and Finance, Vol.4 Issue 4, p.387-410
Gärtner, M., Wellershoff, K.W., 1999. Theories of political cycles: lessons from the American stock market. International Review of Economics, Vol. 46 Issue 4, p.613-630
Havrilesky, T., 1987. A partisanship theory of fiscal and monetary regimes. Journal of Money, Credit and Banking, Vol. 19 Issue 3, p.308-325
Herron, M.C., 2000. Estimating the economic impact of political party competition in the 1992 British election. American Journal of Political Science, Vol. 44 Issue 2, p.326-337
Hibbs, Jr., D.A., 1977. Political parties and macroeconomic policy. American Political Science Review, Vol. 71 Issue 4, p.1467-1487
Huang, R.D., 1985. Common stock returns and presidential elections. Financial Analysts Journal, Vol. 41 Issue 2, p58-61.
Hudson, R., Keasey, K., Dempsey, M., 1998. Share prices under Tory and Labour governments in the UK since 1945. Applied Financial Economics, Vol. 8 Issue 4, p.389-400
Jacobsen, B., Potters, J., Schram, A., van Winden, F., Wit, J., 2000. (In)accuracy of a European political stock market: the influence of common value structures. European Economic Review, Vol. 44 Issue 2, p.205-230
Johnson, R.R., Chittenden, W., Jensen, G., 1999. Presidential politics, stocks bonds, bills and inflation. Journal of Portfolio Management, Vol. 26 Issue 1, p.27-31
Jones, S., Banning, K., 2009. US elections and monthly stock market returns. Journal of Economics and Finance. Vol. 33 Issue 3, p. 273-287
20 Leblang, D., Mukherjee, B., 2005. Government partisanship, elections, and the stock market:
examining American and British stock returns, 1930-2000. American Journal of Political Science. Vol. 49 Issue 4, p. 780-802
McCallum, B.T., 1978. The political business cycle: An empirical test. Southern Economic Journal, Vol. 44 Issue 3, p. 504-515
Nordhaus, W.D., 1975. The political business cycle. Review of Economic Studies, Vol. 42 Issue 2, p.169-190
Orlitzky, M., 2013. Corporate social responsibility, noise, and stock market volatility. The Academy of Management Perspectives, Vol.27 Issue 3, p.238-254
Pantzalis, C., Stangeland, D.A., Turtle, H.J., 2000. Political elections and the resolution of
uncertainty: the international evidence. Journal of Banking & Finance , Vol. 24 Issue 10, p.1575-1604
Santa-Clara, P., Valkonov, R., 2003. The presidential puzzle: political cycles and the stock market. The Journal of Finance, Vol. 58 Issue 5, p.1841-1872
Shum, P.M., 1995. The 1992 Canadian constitutional referendum: using financial data to asses economic consequences. Canadian Journal of Economics. Vol. 28 Issue 4, p. 794-807
Siegel, J.J., 2002. Stocks for the long run, 3rd edition. McGraw Hill, New York
Stovall, R.H., 1992. Forecasting stock market performance via the presidential cycle. Financial Analysts Journal, Vol. 48 No.3, p.5-8
Sturm, R.R., 2009. The ‘other’ January effect and the presidential election cycle. Applied Financial Economics, Vol. 19 Issue 17, p.1355-1363
Sturm, R.R., 2013. Economic policy and the presidential election cycle in stock returns. Journal of Economic Finance. Vol. 37 Issue 2, p.200-215
Summers, L.H., 1986. Does the stock market rationally reflect fundamental values?. Journal of Finance, Vol. 41 Issue 3, p.591-601
