The effect of cross listing on the sensitivity of managerial compensaton to firm performance

1
The effect of cross listing on the sensitivity of
managerial compensaton to firm performance

• Bin Ke
• Pennsylvania State University
• Oliver Rui and Wei Yu
• Chinese University of Hong Kong
• Discussant Raghu Rau
• Purdue University

2
Basic Question

• What is the effect of cross-listing on the
  sensitivity of the pay performance relationship?
• What is the sensitivity of cash compensation to
  firm performance?
• What is the sensitivity of CEO turnover to firm
  performance?
• How does managerial ownership vary with type of
  firm?

3
Data

• Three types of firms in 2003
• A shares Listed only in China
• H shares Cross listed in HK
• Red chips Chinese controlled firms incorporated
  outside China but trade in HK.
• Limit analysis to firms that earn gt50 of their
  revenues from mainland China.
• Restrictions on firm size, values for assets etc.

Explaining allocation to equity
4
Hypotheses

• Investor protection increases from A shares to H
  shares to Red Chips.
• Red chip sensitivitygtH share sensitivity gt A
  shares sensitivity to performance
• State controlled firms are different from public
  firms
• Different motivations of the state (political,
  employment, equity)
• Results for state-owned firms perhaps only in Red
  chips.

Explaining allocation to equity
5
Methodology

• Regressions
• Ln(Cashpay) on performance, H, Red chip,
  interaction variables, control variables
• Equity ownership on H, redchip, control variables
• Cox proportional hazard model for CEO turnover

Explaining allocation to equity
6
Results

• Cash compensation sensitivity
• Similar across A, H and red chips for non-state
  controlled firms
• For state controlled firms, Red chips and H are
  more sensitive than A shares for performance
  measured by returns but not ROA
• Managerial ownership
• Non-state controlled
• No differences for CEO
• Non-CEOs Significantly higher ownership for H
  and marginally higher for Red chips
• State controlled
• Red chips have significantly higher equity
  ownership
• Managerial turnover
• Higher managerial turnover sensitivity for
  non-state controlled firms to H-shares than
  A-shares but not red chips

Explaining allocation to equity
7
Comments ?

• Good well-written paper
• Looks at one country three systems to develop a
  set of logical testable hypotheses
• Distinguishes between state and non-state owned
  firms and develops differential hypotheses on
  them.
• Investigates a new area how compensation issues
  can be solved through cross-listing and other
  bonding mechanisms.

8
However ?

• Do you really need to distinguish between
  state-controlled and non-state controlled firms?
• The regressions in tables 2,3,4 dont look very
  different either in the magnitude or their
  significance.
• What about one regression with a dummy variable
  for state ownership or even the proportion of
  state ownership?
• Or tests to show they are different.

9
However ?

• If state-controlled firms and non-state
  controlled firms are different animals (as the
  authors suggest) then the cross-listing decision
  is endogenous only for the state-controlled
  firms.
• This explains why the authors find some increased
  sensitivity for H-shares for non-state controlled
  firms.
• They dont find any for state-controlled firms
  because the decision to cross list is driven by
  different variables for these firms.

10
However ?

• Are H-shares, A-shares or red chips really
  systematically different in the degree of
  corporate protection?
• Remember judicial decisions in HK are not
  enforceable on the mainland.
• Authors throw out firms that do not do a majority
  of their business with the mainland but perhaps
  these are the very firms that will show a
  difference.

11
However ?

• A-shares are also different animals whether they
  are controlled by local governments than central
  governments.
• Add Distance to Beijing as a controlling variable
  
• Other distinguishing variables
• Did the firm issue ADRs?
• Did the firm have a big 4 auditor?

12
However ?

• Managerial ownership is not really testing the
  sensitivity of option compensation to performance
  (as the authors claim).
• This part of the paper does not really fit that
  well
• Raises issues of causality
• When was the ownership established?

13
However ?

• Some coefficients are puzzling in the
  interpretations.
• Table 2 says that RET is important in determining
  cash compensation in state controlled firms
• Table 4 says that ROA is important in determining
  turnover in state controlled firms
• Signs switch between state and non-state
  controlled firms.