Dr. Qi Deng is an associate professor in finance with IBSSXJTLU since May 2014. He co-founded Altobeam, an award-winning technology company in 2007, and he led its financial and business development activities until September 2016 when he founded Cofintelligence Technology Ltd., a financial technology firm that specializes in machine learning-based quantitative modeling and algorithms trading, and he serves as its managing director and chief scientist.
Since the establishment of Chinese IPO markets in 1991 at the Shanghai Stock Exchange
(SHSE), Chinese IPOs have experienced severe underpricing, with main reasons being policydriven,
regulation-induced quota system, offer size, P/E ratio cap, and long delay from pricing to
listing. The launch of the ChiNext Board on October 30, 2009 represents Chinese authority’s
initial efforts to make new stock issuance and thereafter stock trading more transparent and
market-driven. The ChiNext board quickly becomes China’s first board in which a Chinese-style
bookbuilding process is utilized to price the IPOs listed on ChiNext board ever since its
inauguration, and thus provides an excellent opportunity to investigate Chinese IPO pricing and
underpricing in a more market-oriented environment.
The Chinese-style bookbuilding is unique and it includes a 2-phase process. During the
1st phase (offline stage), institutional investors are invited to bid for new issue shares, but the
underwriter has no discretion in share allocation, and a prorated and later on a lottery mechanism
is utilized in allocating shares to the institutional investors. The underwriter determines the issue
price (offer price) during the bookbuilding stage. The retail/individual investors then go through
a lottery system to bid for allocations in the 2nd phase (online stage). Once the IPO pricing and
allocation process is complete, the issue is in the quote, waiting for its turn to get listed.
To our best knowledge, there is no thorough research on what the determinants are in IPO
pricing and underpricing specifically for the ChiNext board. This dissertation conducts empirical
tests on all 355 ChiNext IPOs initiated at the Shenzhen Stock Exchange (SZSE) between October
30, 2009 and December 2012 in order to find out the determinants for ChiNext IPOs’ offer prices,
the 1st listing day opening price returns, the 1st listing day closing price returns (the first day
initial underpricing), and the 21st trading day price returns (the first month initial underpricing).
The key findings are: in the issue pricing stage, the institutional investors base their bids
mainly on the issuer’s post-issue financials (especially post-issue asset per share), the underwriter
largely prices the issue based on the institutional investors’ biding prices (especially the weighted
average price); the 1st day opening price return is mainly determined by the pre-issue allocation
- ii -
ratio among institutional investors and the issue size; the 1st day closing price return is mainly
determined by the pre-issue allocation ratio among institutional investors, the listing day market
condition, and the issue size; the 21st day closing pricing return is mainly driven by the pre-issue
allocation ratio among institutional investors, the listing day market condition, the issue size, the
delay between pricing and listing dates, the 1st day trading value,