Research Journal of Management Sciences _____________________________________________ISSN 2319–1171 Vol. 4(2), 1-9, February (2015) Res. J. Management Sci. International Science Congress Association 1 Empirical Study on Effects of the Lok Sabha Election on Stock Market Performance (BSE SENSEX) Kumar Deva B., Sophia Sharon and Jucunda Evelyn Maria2 The University of Michigan, Dearborn MBA in VIT Business School, Chennai, INDIA VIT Business School, Chennai, INDIAAvailable online at: www.isca.in, www.isca.me Received 4th January 2015, revised 28th January 2015, accepted 5th February 2015 Abstract This study is mainly focused on effects of 16thLok Sabha election on stock market performance in India. Various factors have been analysed which affect the stock market performance during election period. The study has found that there is significant relationship between Lok Sabha election and Stock market performance. People’s sentimental analysis about the companies and also Portfolio optimization for the companies which are listed in BSE SENSEX index have been done during election time. This study examines that Lok Sabha election affects the stock market performance and also company’s endogenous and exogenous factors are affected. In addition to, Election result is also one of the factors of an investment decision which has proved in this study by the way of conducting an Event study. Keywords: Elections, sentimental analysis, event study, abnormal returns. Introduction History has demonstrated that stock market (financial system) plays important role in economy for economic growth. Stock market is also reflects the country’s status. Now a days, stock market has been impacted by many factors which are political, weather condition etc., we are in technological world where news could be spread all over the world within a short span of time. Because of this, every news could be reached to the public so that investors could analyse news which is related to stock market. So stock market fluctuation is based on news. The news which is related to economy, political events weather conditions and relationship between countries could impact on stock market fluctuations. We have taken an event which has been an important reason of stock market fluctuations in India. The event is Lok Sabha Election in India 2014. Political events usually have great impact on the stock market. In many cases, the stock market fluctuates because of political announcements such as regulation promulgation, law amendments and national elections. Moreover, this year Lok Sabha election (2014) is most important election in Indian political history. Congress was ruling government of India for last ten years. Nevertheless, Congress party did not run a government properly so that people need a new government which would be helpful for economic growth. By which time Mr Narendra Modi waves have spread throughout the country. Narendra Modi was Chief Minister of Gujarat. He has done many things which helped for economic growth of Gujarat. So he has got a good reputation across the country. This news spread all over the country then all over the world. Indian investors and also foreign investors have believed that there would be chance for economic growth in India, if Modi becomes Prime Minister. Indian stock market has positive impact when Narendra Modi was selected for prime ministerial candidate. This study discusses the impact of Lok Sabha elections on stock market. BSE SENSEX Index and Companies have been taken for the analysis. Event study has been conducted in this paper. The event is 16thloksabha election result date (16th may 2014). Sentimental analysis (Rstudio) has been done for comparing market movements and investor’s sentiments by using social network Twitter. Company’s exogenous and endogenous factors have been analysed. Measuring abnormal returns for firms on particular events like elections, acquisitions, stock market crash and so on has been an interesting area of study to researchers in the west and also in India. The most common measure of abnormal returns across the globe is the event stud methodology which measures the returns for specific windows. The abnormal returns for a particular event may be due to various reasons. One such reason is the information asymmetry and rumours that spreads across the market about a particular event. In recent years after the introduction of internet, the social media has become one of the most powerful medium of information dissemination. Researchers have tried to capture the use of social media and its information in predicting events such as tsunami, movies success, election polls results and so on. Thus prediction of events and results from the information in social media is becoming an upcoming and attractive topic to researchers across the world. The recent improvement in this field is capturing the sentiments of investors through social media and predicting the stock Research Journal of Management Sciences _________________________________________________________ISSN 2319–1171Vol. 4(2), 1-9, February (2015) Res. J. Management Sci.International Science Congress Association 2 market behaviour using the investor sentiments. Twitter is one of the social media which is highly accepted in the financial community. Messages from twitter, known as tweets can be easily accessed through application programming interface (API). Many sub forums in twitter has been started recently like Stock twitsand Tweet Trader which acts as a platform for discussion among the investors. Researchers have tried to capture the stock market behaviour using the investor sentiments derived from twitter using Google Profile of Mood States, Opinion Finder and so on2,3,4. Twitter was established in 2006 and since then the number of people and firms joining twitter have been increasing drastically. Every day around 65 million tweets are posted per day with 750 tweets each second. Though researchers in the west have started to exploit the information from twitter, in India it is still an unexplored area. Indian population is the second in the world in the usage of internet and thus the importance of market sentiments from social media cannot be undermined. This study takes a different stand from literature as it studies the influence of twitter sentiments on the abnormal returns of firms on a particular event. Unlike the literature which predicts the stock behaviour of tomorrow using today’s tweets, this study, studies the influence of twitter tweets around the event announcement on the abnormal returns and cumulativeabnormal returns around the announcement. The eventconsidered in this study is the announcement of the16THLoksabha election result date (16th may 2014). Thus the objectives of this paper are multiple: Objectives of the study: i.To analyse the twitter sentiments from twitter tweets the behaviour of stock market around the event specified. ii. To identify the changes in the abnormal return of stocks due to specified event. iii. To identify the endogenous and exogenous factors that contributes significantly towards specified event. Review of literature: Ling-Chun Hung (2011) has described that History has demonstrated the fact that politics and economy are intertwined. Presidential election is considered the most powerful political event. This study examines three Taiwanese presidential elections after the year 2000 in order to investigate the existence of short-term (bull-run election) and long-term (election cycle) effects as well as the myth regarding the market favouring a particular party (The Kuomintang). The findings indicate that there is an election cycle in the Taiwanese presidential election, but there is no proof for a bull-run election and no evidence for the market’s preference. Ling-Fang Liu (2007)examines that stock market is fluctuating by election results. In addition, there is positive reaction over the stock exchange after 15 days and before 15 days which has been proven by conducting event study. Moreover, other financial and political factors have been found to play an important role in influencing the return pattern around presidential elections. Ray M. Valadez, Marshall D. Nickles (2009) focus on the relationship between political parties holding presidential office and SandP 500 performance, Volatility and risk. The research encompassed a several steps. First, the specific years within presidential cycle were evaluated for price change over time. Next, Observation were to determine if there were any relationship between political party in office and historical stock market performance from the post- world war-II period to the present. Moreover, there is four year presidential election cycle in the country but the result shows that there is no significance between political parties and historical stock market performance. Angela, Kithanji and Wilson Ngugi (2007) examined that analyses the performance of Nairobi Stock Exchange before and after the last four general elections in Kenya. The study focused on the NSE performance before and after the 1992, 1997, 2002 and 2007 general elections. The study results indicate that the NSE performance was influenced by the political activities and expectations around the election period in the short-term. In addition, the study also reveals that the first two years after the general elections the NSE performed better than the last two years before the next general elections. The poor performance before the election could be attributed to investor anxiety and panic associated with pre-election period. Wing-Keung Wong and Michael McAleer (2007) analysed the impact of Presidential elections on stock prices with reference to the USA. Using spectral analysis and the EGARCH Intervention model, the results suggests that since 1965, the US stock market has experienced several Presidential Election Cycles. Nicholas Chen (2004)10foresees that in the almost four decades from January 1965 through to December 2003, US stock prices closely followed the four-year Presidential Election Cycle. The empirical results suggest that the Republican Party may have greater cause to engage in active policy manipulation to win re-election than their Democratic counterparts. The existence of the Presidential Election Cycle shown in the paper may constitute an anomaly in the US stock market, which could be useful for investors. Yi-Hsien Wang, Mei-Yu Lee and Che-Yang Lin (2008)11analysed the U.S presidential elections are a major event that takes place every 4 years and affects economies all over the world. Election results may influence corporate performance by changes in government policies such as spending and tax changes. Further, specific sectors might gain or suffer from sector-specific governmental policies. It can be seen that elections do affect stock markets in a certain direction, depending on both the individual president’s themselves as well as the general policies the winning parties will undertake and the possible effect of those policies on market reactions and sentiments. In addition Research Journal of Management Sciences _________________________________________________________ISSN 2319–1171Vol. 4(2), 1-9, February (2015) Res. J. Management Sci.International Science Congress Association 3 to this, we must look at the other economic factors in the environment at the particular time and place in order to make a good assessment of the market direction. Effects of post elections must also be taken into account, such as the so called fiscal cliff for the 2012 elections. David Leblang and Bumba Mukherjee12analysed the impact of Presidential elections on stock prices with reference to the USA. The empirical results using spectral analysis and the EGARCH Intervention model found that, since 1965, the US stock market has experienced several robust and quantitatively important Presidential Election Cycles. Sentimental analysis review: Bollenand Mao study analyzed the public’s emotional state over a month period by using a term based emotional rating system known as Profile Of Mood States (POMS). Sentimental analyses is performed for all public tweets broadcasted by twitter users between aug 1 and dec 20 2008. The results were compared to fluctuations recorded b stock market and crude oil price indices and major events in media. The results fund that the events in the social, cultural and economic sphere do have a significant, immediate and highly specific effect on the various dimensions of public mood. Bollen, Mao and Zheng13 investigated whether the measurement of collective mood states derived from large scale twitter feeds were correlated to the value of DJIA overtime. The study used opinion finder and GPOMS to measure variations in the public mood from tweets submitted to the twitter service from feb to dec 2008. Granger Casuality analyses were used to correlate DJIA values to GPOMS and OF values for the past n days. Next the study used self-organizing fuzzy nueral network model to test the hypothesis that the prediction accuracy of DJIA prediction models using measurements of public mood. A correlation of mood time series was drawn between GPOMS and OF and was found that certain mood dimensions of GPOMS partially overlap with OF. Zhnag, Fuehresand Gloor tried to predict stock market indicators such as DowJones, NASDAQ and SandP500 by analyzing twitter posts. They analyzed the positive and negative moods of the masses of twitter for a period of 6 months and compared it with stock market indices. The study investigated the emotions of the tweets under three different baselines: number of tweets per day, number of followers per day and number of re-tweets per day. The study correlated the ratio of emotions with the indices returns for the day t+1. The results were surprising as it found that people start using emotional words such as ‘hope’, ‘fear’ and ‘worry’ in times of economic uncertainty, independent of whether they have a positive or a negative context. Thus when the emotions on twitter fl high, the Dow goes down the next day and when people have less hope, fear and worry Dow goes up. Hypothesis: H1: There is a significance relationship between Loksabha election result (2014) and stock market. H2: There is a significance changes in the abnormal return of companies H3: Endogenous and exogenous factors contribute significant towards a specified event. Methodology Sample selection: Therefore 30 companies of the BSE SENSEX are selected for the study. The companies selected were expected to fulfil the following criteria: i. The companies should be listed in Bombay Stock Exchange (SENSEX). ii. The companies should have market data for (-14,+14) days around the specified announcement. iii. The companies should be active in twitter. Companies Name and Script Code: The 30 companies and their script code in BSE is listed below. Table-1 Company and scrip name Company Name Scrip Code Axis Bank 532215 Bajaj Auto 532977 Bharat Heavy Electricals 500103 Bharti Airtel 532454 Cipla 500087 Coal India 533278 Dr. Reddy's 500124 GAIL 532155 HDFC Bank 500180 Hero Motocorp 500182 Hindalco Industries 500440 Hindustan Unilever 500696 Housing Development Finance Corporation 500010 ICICI Bank 532174 Infosys 500209 ITC 500875 Larsen and Toubro 500510 Mahindra and Mahindra 500520 Maruti Suzuki 532500 NTPC 532555 Oil and Natural Gas Corporation 500312 Reliance Industries 500325 SesaSterlite Ltd 500295 State Bank Of India 500112 Sun Pharmaceutical 524715 Tata Consultancy Services 532540 Tata Motors 500570 Tata Power 500400 Tata Steel 500470 Wipro 507685 Classification of Companies: The 30 Companies have been classified by industry. Research Journal of Management Sciences _________________________________________________________ISSN 2319–1171Vol. 4(2), 1-9, February (2015) Res. J. Management Sci.International Science Congress Association 4 Table-2 Number of companies in each sector Classification of Company No of Companies Automotive 5 Banking 4 Conglomerate 2 Consumer goods 1 Electrical equipment 1 Housing Finance 1 Information Technology 3 Iron and Steel 1 Metals and Mining 2 Oil and gas 3 Pharmaceuticals 3 Power 2 Steel 1 Telecommunication 1 Figure-1 No of companies in each sector Twitter Sentimental Analysis: The twitter tweets were collected for the 30 companies in the sample for 30 days around the election announcement. As a developing economy, tweets for Indian companies will be comparatively less when compared to the companies in the developed economies. The number of tweets collected for each company is shown below. No. of Tweets and Company Name: The tweets were collected manually for each company. The collected tweets were subjected to sentimental analysis throughR-Studio.   \n  \r     \n\r\r  \n\r  \r \r \n\r    \r  \n !\r"  #  $\r\r %\r%   &  '"\n\r  ' (\r $\r\r !\r\r \n\n