Do Stock Prices Conform to an Absolute Price Level?

Abstract

We examine nominal and real stock prices and the sequential price pattern of stock dividends, stock splits and Initial Public Offerings (IPOs). Using data for all firms listed in Sweden over the one hundred and twelve year period 1900-2011, we show that the average stock price has been fairly stable over time except for two decades in the beginning and end of the 20 th century. Inclusion of these periods yield a decline over time which is generally consistent with the drop in price levels found by Chittenden et al. (2010). In a multivariate setting, the frequency of stock dividends, stock splits and IPOs are positively related to the frequency for these events the prior year and recent market return. In further tests of the price change we find a positive relationship between the price difference to the median price for stock splits and stock dividends and a positive relationship to the market return for stock splits. We conclude that the primary reason for an action such as a stock split, is to fit the "norm" price of the market. * We would like to thank the discussants Mike Staunton at the 2010 EurHISTOCK Conference (Cambridge), Omid Sabbaghi at the 2011 EFA Conference (Savannah) and seminar participants for their many helpful comments. We also thank Mikael Segerlund for his excellent research assistance. Financial support from the Swedish Securities Dealers Association is gratefully acknowledged. The usual disclaimer applies. ** Corresponding author. 1 Do Stock Prices Conform to an Absolute Price Level? AbstractWe examine nominal and real stock prices and the sequential price pattern of stock dividends, stock splits and Initial Public Offerings (IPOs). Using data for all firms listed in Sweden over the one hundred and twelve year period 1900-2011, we show that the average stock price has been fairly stable over time except for two decades in the beginning and end of the 20th century. Inclusion of these periods yield a decline over time which is generally consistent with the drop in price levels found by Chittenden et al. (2010). In a multivariate setting, the frequency of stock dividends, stock splits and IPOs are positively related to the frequency for these events the prior year and recent market return. In further tests of the price change we find a positive relationship between the price difference to the median price for stock splits and stock dividends and a positive relationship to the market return for stock splits. We conclude that the primary reason for an action such as a stock split, is to fit the "norm" price of the market.