دانلود رایگان مقاله انگلیسی موانع قیمت ارقام در بازار سهام ایالات متحده - SSRN 2018

Abstract

This paper examines the existence of round number price barriers in the U.S. stock market. I show that stock prices clusters around multiples of $10 as a result of the price barriers. The price barriers results in abnormal future return pattern; a long-short portfolio formed around the barrier held for a week produces 17 basis point weekly return (8% annually). Such abnormal return does not change in magnitude after controlling for varies factors. I present supportive evidence for the left-digit-bias barrier channel.

4. Potential Channels

I propose the left-digit-bias induced psychological barriers as a channel that explains the observed price clustering and the abnormal return. The channel implications are consistent with the empirical results. After discussing the left-digit-bias channel I’ll look at other potential channels and whether they fit with the empirical observations.

Left-Digit-Bias Induced Barriers

This channel begins with assuming that investors suffers from left digit bias; Specifically, the tens digit is more salient than the integer digit to an investor. Put it in other words, investors mentally put stocks into bins base on their tens digit, where stocks with price $[10,20) is in one bin and stocks with price $[20,30) is in another bin, etc. A $20 stock appears to be substantially higher in value than a $19 stock because it is in a higher value bin (while a $17 stock appears similar in value to a $16 stock, being from the same bin). Such mental binning of stocks results in a discontinuous jump in the perceived value of a stock at multiples of $10, thus creating stock prices barriers at tens ($10, $20, $30, etc), making it harder for stock price to move from $19 to $20 (and vice versa).

The first consequence of barriers at multiples of $10 are clustering of stock prices. Looking at the empirical distribution of prices in figure 2, we observe a discontinuous drop in frequency at $10, consistent with barriers blocking price movements from above. The second consequence is that barriers prevents information from incorporating into prices. Imagine good news arrives for a stock at $19, and normally such news would push its price above $20. Due to the barrier at $20 however, investors are reluctant to buy with a price above $20. This creates a temporary underpricing for the stock. The good news eventually prevails and push the price through the $20 barrier. Now there is no long a barrier above and the underpricing are being corrected, resulting in an abnormal high return of stocks just above $20. Again this fits well with the empirical observation that stocks with integer digit 9 earns an abnormal low return while stocks with integer digit 0 earns an abnormal high return.