The impact of speculation upon volatility and market e ciency: The badla experience on the BSE

On 12 March 1994, SEBI imposed new norms on trading in the Bombay Stock Exchange, and the e ective consequence of this has been an elimination of badla, a form of forward trading. Without badla, the role of speculative traders on the BSE is diminished. This paper sets out to measure the impact of this elimination of speculative trading upon volatility on the BSE. We explore, and criticise, a variety of di erent methods of arriving at a conclusion on this question, and present an estimation strategy which exploits the unique opportunity to view this episode as a natural experiment. Our examination of daily unsystematic risk, which takes the value of 3% in our sample on average, reveals that badla diminishes it by roughly 0.25 percentage points. The statistical signi cance of this estimate is weak, especially in the light of a qualitative argument suggesting that this estimate is biased upwards. Working with weekly returns data, badla seems to have no impact upon unsystematic risk. On the subject of market e ciency, we nd that badla is slightly bene cial for short{horizon market e ciency: the non-forecastability of daily returns of A companies has worsened in the year following 12 March 1994. This e ect is concentrated in the short horizon; the degree of forecastability of weekly returns has actually diminished slightly in the year after 12 March 1994. This le is http://www.cmie.ernet.in/ ajayshah/PROSE/ARTICLES/badla.ps.gz The estimation strategy used here evolved in email conversations with Venkat Eleswarapu. Mailing address: Centre for Monitoring Indian Economy, 110-120 Kaliandas Udyog Bhavan, Prabhadevi, Bombay 400 025.