Investor-Owned Firms: Corporate Governance

Abe and Shimizutani, 2007, Employment policy and corporate governance: an

empirical comparison of the stakeholder and the profit-maximization model

1. The question the article tries to answer

How ownership structures affect firms decisions as to how to reduce labour

costs when firms face excess employment.

2. The theory behind the hypotheses

The stakeholder view of the firm vs the neoclassical view of the firm as a
profit-maximizer;

Stakeholder view: not only shareholders and managers but also other
stakeholders may play significant roles in the decision-making process
- Theoretical analysis of the stakeholder model Tirole (2001)
- Both theoretical and empirical investigation Blair and Roe (1999)
- Debates on the issue originate in 1976 Drucker

- Other works: critical view of stakeholder models as sustainable

institutional arrangements Jensen and Mecking (1979), Fama and Jensen
(1983a, 1983b) and Jensen (2000).

The results of the empirical analysis are consistent with the stakeholder view.

The theory behind the hypotheses cond.

Recent theoretical studies have elaborated of stakeholder view Pagano and

Volpin (2005a)

The degree of institutional investor protection is a function of the distribution

of financial wealth or company shares Perroti and Thadden (2005) and
Pagano and Volpin (2005b)

3. Theoretical hypotheses tested

Empirical studies support theoretical conclusions:

- Falaye et at. (2006) found that the labour influence in many US firms may
be stronger than is generally appreciated
- Fukao and Morita (1997) and Morck and Steier (2005) compare governance
structures in Japan from an international perspective

- Aoki et al. (1994) model where Japanese long-term employment system

and the main bank system are complementary so that efficient source allocation
is achieved Contingent governance theory

Theoretical hypotheses tested cond.

Unique survey directly asked managers about the degree of excess

employment and the way they were planning to deal with the problem.

Merge the results of the survey with other company information (ownership
structures, board compositions and financial data)

Examine the effect of changes in corporate governance on employment

downsizing measures and corporate governance.

Further theory and hypothesis

Standard Corporate Governance Theory

- Based on principleagent models: the role of inside and outside director s is
different (Hermalin and Weisbach 2003) and Murphy (1999)
Insider-Outsider theory (in macroeconomics literature) Lindbeck (1993)
Both of the theories predict that companies with fewer outside directors are
more inclined to pursue the interest of their employees rather than those of
other stakeholders such as shareholders.

4. Empirical specification and data

Three data sets:

1.

Survey data at firm-level , the Annual Survey of Corporate Behavior, 2001 =>
perceived excess or shortage of employment, companies which needed to
decrease labour cost and eight measures to do so (reduction in wages or
working hours, layoffs, voluntary retirement, etc.);

2.

Directors Data (Yakuin Shikihou) published by Toyo Keizai Shimpo-sha data

set with detailed information on the board members of all Japanese listed
companies => differentiate between insider directors (promoted from among
employees) and outsiders;

3.

Financial statements of all listed firms, from Nikkei NEEDS database.

Sample of 317 manufacturing firms with excess employment;

Variables: number of employees, proportion of insiders among board

members, board size, natural log of total assets, etc.

5.1. Estimation method

Multivariate probit model to investigate how a larger share of outside (inside)

directors affects which measures firms with excess employment adopted to cut
labor expenses.

For each of the 8 measures below, use a dummy variable which takes value 1 if
the firm has implemented it or will implement it and 0 otherwise:
(1) reduction in bonuses;
(2) reduction in wage rates;
(3) reduction in managers salaries;
(4) reduction in executive payments;
(5) reduction in working hours;
(6) reduction in hiring new employees;
(7) expansion of layoffs; and
(8) introduction of voluntary (early) retirement.

5.2. Estimation method

The model consists of 8 equations:

where i: firm index and j : equation index that corresponds to the measure of labour cost reduction,
j = 1, 2,.., 8.
: dummy variable for bonus reduction;
: dummy for wage reduction;
dummy for reduction in managers salaries;
: dummy for reduction in executive payments;
: dummy for reduction in working hours;
: dummy for reduction in hiring new employees;
: dummy for layoffs;
: dummy for introduction of voluntary (early) retirements;

: the ratio of insiders on the board of directors;

: firm characteristics (ROA, the number of employees, etc.);
: ownership structure.

6. Main findings

6. Main findings

The Insider Ratio has a negative effect on the adoption of measures that are
costly for incumbent employees (i.e. bonus reduction, wage reduction) and a
positive effect on measure that are costly for new employees (i.e. reduction
in new hiring).

Among firms suffering from excess employment, those with a greater

proportion of outside directors are more likely to lay off staff or resort to
voluntary early retirements, while boards mostly consisting of insiders are
more inclined to reducing new hiring.

These results are consistent with the stakeholder view of the firm suggesting
that inside directors are more concerned with protecting the interests of
employees than with profit-maximization as assumed by neoclassical theory.

7. Critique

The composition of boards of directors is endogenously determined which can

result in a bias;

This is due to the fact that data on employment policy is cross-sectional, plus
it is difficult to find good instruments for board composition;

To correct for endogeneity, use of an independent lagged variable: the insider

ratio of 2001 is replaced with that of 1996;

However, there may be unobservable omitted variables such as firm culture

that determines both board composition and labor cost reduction measures;

Firms estimated the gap between the current and the optimal level of
employment; optimal level of employment may be inconsistent among firms.

Conclusions