Odd Lot Trades: The Behavior, Characteristics, and Information Content, Over Time

Date01 November 2014
Published date01 November 2014
The Financial Review 49 (2014) 669–684
Odd Lot Trades: The Behavior,
Characteristics, and Information Content,
Over Time
Hardy Johnson*
Kansas State University
I investigate the role odd lot trades play in equity markets, and howthis role changes over
four periods: 2005, 2007, 2010, and 2012. In each of these years, I examine the determinants,
price contribution, and characteristics of odd lot trading. I find that odd lot proportions are
increasing, but the determinants of the proportions remain consistent. I find that odd lot
transactions contribute to price formation, this contributionis in excess of the odd lot proportion
of volume, and is increasing over time. An intraweek pattern of odd lot trading exists with
Monday having the highest proportions.
Keywords: odd lot trades, market microstructure, price discovery, day-of-the-week patterns
JEL Classifications: G10, G14
1. Introduction
Until recently, odd lot transactions (trades for fewer than 100 shares) had not
been extensively examined by capital markets researchers. It was assumed odd lot
*Corresponding author: College of Business Administration, 110 Calvin Hall, Kansas State University,
Manhattan, KS 66506; Phone: (785) 532–6992; Fax: (785) 531-6822; E-mail: bhjohnson@k-state.edu.
I would like to thank the editor,Robert Van Ness, and the anonymous referees for constructive and detailed
C2014 The Eastern Finance Association 669
670 H. Johnson/The Financial Review 49 (2014) 669–684
traders were individuals, and therefore uninformed and contributed little, if anything,
to price formation. However, O’Hara, Yao and Ye (2014) show that, in a sample of
120 stocks from 2008 to 2009, odd lot trades comprise roughly 8% of total volume
and 20% of all trades and odd lot transactions are contributing to price discovery.
They also show that odd lot proportions of volumeand trades are increasing over their
two year sample of 2008–2009. This study investigates the role odd lot trades play in
equity markets, as well as how this role changes from 2005 to 2012. I seek to answer
whether increasing odd lot proportions is a secular trend by using a longer time frame
than that used by O’Hara, Yaoand Ye (2014). Also, I examine the information content
of odd lot transactions to determine if odd lots are playing an increasing role in price
formation as their proportions increase.
From a larger data set that contains all trades in all stocks that execute on
NASDAQ, my results show that odd lot proportions and their informativeness are
increasing during four distinct periods: 2005, 2007, 2010, and 2012. I investigate
several characteristics of odd lot trades to determine whether this is due to fluctuations
in market variables that have been shown to be determinants of odd lot proportions,
as suggested by O’Hara, Yao and Ye (2014). Because many stock characteristics
display an intraweek pattern, I also document the characteristics of odd lot trading
on a day-of-the-week basis.
I define odd lots as trades of less than 100 shares (including those that are a
partial execution of 100+share orders). I do not differentiate between odd lot trades
that are submitted as odd lot orders and odd lot trades that are submitted as part of
a 100+share order, but execute in more than one trade with one of the trades being
for less than 100 shares (e.g., a 250 share order that executes as two orders: one for
200 shares and one for 50 shares). There is limited research on odd lot transactions
as trades of less than 100 shares are not reported in the NYSE Trades and Quotes
(TAQ) database. Trades of more than 100 shares, but not in 100 share increments (for
125 shares, for example), are included in TAQ.
Although odd lot transactions can be submitted by either institutional or indi-
vidual traders, a number of previous studies assume odd lot trades are individuals’
transactions. Ritter (1988) and Dyl and Maberly (1992) use odd lots trades as a proxy
for individuals’ trades in studies that investigate the Turn-of-the-Year Effect. Simi-
larly, Lakonishok and Maberly (1990) use odd lot trades as a proxy for individuals’
trades in their study on the Weekend Effect.
In contrast, recent studies suggest that odd lot transactions are not necessarily
trades by individuals. Using a database of NASDAQ transactions for 120 stocks
(including odd lot transactions), O’Hara, Yao and Ye (2014) find that roughly one-
quarter of all trades initiated by high frequency traders (institutional traders) are for
less than 100 shares.
In 2012, odd lot transactions are approximately 30% of the average stock’s
transactions, accounting for approximately 12% of volume. It is nontrivial that 12%
of volume and 30% of trades are omitted from previous studies that use the TAQ
database (as TAQ does not contain odd lot trades).

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT