| dc.description.abstract | This study looked at the stock market behaviour around national elections in Kenya. Two periods were chosen namely 1997 and 2002 elections mainly because of data availability. Previous empirical research has shown that share prices respond to events. For instance, Richard (2006) cites national elections as one of the event classes that is anticipated in share price movements. A country's politics can exert significant influence on its income distribution and prosperity. Despite the importance attached to election effects, most previous studies have only concentrated on the developed countries but no scholarly work has been done in Kenya. It is therefore from this stand point that this study was warranted and therefore addressed this gap.
This study was based on an event study methodology of the stock market behaviour around the national elections of 1997 and 2002. The focus was on the NSE 20 Share index consistuent counters. These were chosen as they represent 80% of the trading volumes at the Nairobi Stock Exchange. A 60-day trading window around national election dates was chosen and the market model approach based on the 250-day daily returns benchmark model by Brown (1985) was employed. Previous studies have shown that event studies based on market model and standard parametric tests are well specified under a variety of conditions. Hence, this method provided the framework for this study.
The study's findings mirror what has been observed in previous studies. The stocks react strongly to elections outcome and
temporarily elevated levels of volatility are observed. Share prices go down before the election date and start rising thereafter. Abnormal returns shift steeply downwards after the election date and start increasing thereafter before settling to a new equilibrium. However, in both election years the abnormal returns were not found to be statistica lIy sig n ifica nt.
Several limitations were encountered which included lack of data availability, problems of survivorship bias coupled with low liquidity or infrequent trading. This made it difficult for share price information to be gathered and meaningful conclusions to be derived.
This study has mainly two implications for the investors namely compensation for risk and trading strategies. One possible strategy that can be employed from the findings is to use a contrarian strategy before the election event and a momentum strategy after the election event. Both of these strategies have been covered in detail in the literature review section. | en |
