Corporate Governance, and the Returns on Investment Klaus Gugler, Dennis C. Mueller and B. Burcin Yurtoglu University of Vienna, Department of Economics BWZ, Bruennerstr. 72, A-1210, Vienna 1
Considerable interest recently in differences in corporate governance systems. Ten years ago -- German and Japanese systems were thought to be better. Today, many think US & UK (Anglo-Saxon) systems better. 2
Focal point of literature has been on the extent to which corporate governance institutions align the interests of managers and shareholders. Weak corporate governance systems allow managers to pursue their own goals. Much of the literature has emphasized the importance of growth or empire building as a managerial goal. 3
Previous Tests Dividend payouts Tobin s q Both measures of performance have difficulties. When investment opportunities differ, so too should optimal dividend payments. Tobin s q is a measure of average performance. 4
Tobin s q equals the return on total capital divided by the cost of capital. When testing for the existence of agency problems, one wants to compare the returns on investment to the cost of capital. Our paper employs such a measure by estimating marginal qs for firms -- ratios of returns on investment to costs of capital. 5
In testing for the presence of agency problems, we focus upon two sets of institutions The corporate governance structure of a country as defined by its legal system The ownership structure of a company 6
In our empirical work we measure a firm s average return on investment (r) as a ratio to its cost of capital (i), a ratio that we define as qm, qm=r/i. Since the average return on capital should be greater or equal to the marginal return on capital, we predict for a firm, which maximizes shareholder wealth, qm 1. 7
Main Hypotheses (1) Legal Institutions and Returns on Investment A strong corporate governance system aligns managerial and shareholder interests. Managerial and shareholder interests are more likely to be aligned in countries in which it is easy for shareholders to monitor managers, and initiate proxy fights or hostile takeovers. 8
We use the LLSV (1997, 1998) categorization of the legal environments of countries as a measure of the strength of a country s corporate governance system. Hypothesis 1: For companies located in countries with strong corporate governance systems, qm > 1. 9
Weak corporate governance systems allow managers to pursue their own goals at the shareholders expense. Where corporate governance systems are weak managers pay out less in dividends and retain larger fractions of their cash flows to pursue their own goals and invest too much. This reasoning leads us to expect qms < 1, in countries with weak corporate governance systems. 10
Some firms have attractive investment opportunities and limited financial resources. For these companies no conflict between managers and shareholders over dividend and investment policies exists. Hypothesis 2: For companies with limited investment opportunities that are located in countries with weak corporate governance systems, qm < 1. 11
(2) Legal Institutions, Ownership Structures and Returns on Investment We define five ownership categories: Family-controlled Firm-controlled, Finance-controlled State-controlled Dispersed ownership 12
Separation of ownership from control Hypothesis 3: In countries with strong corporate governance systems, companies with widely dispersed shareholdings have lower qms than the other companies in their country group. Hypothesis 4: In countries with weak corporate governance systems, companies with widely dispersed shareholdings have higher qms than the other companies in their country group. 13
No (simple) predictions concerning other ownership categories: family-control: MSV (1988) observe a nonlinear relationship in countries with weak protections it may lead to expropriation or over-investment 14
Control by a financial or non-financial firm: can have several implications on investment performance: + Those in control of any firm A, even if they are empire builders, will want the managers of any firm B that they control to maximize profits 15
- Families behind these pyramids are empire builders who sacrifice profits at all levels of the pyramid - The pyramid s size makes careful monitoring of lower-level firms difficult - The performance of lower level firms is sacrificed to benefit the parent firm at the top of the pyramid 16
qm I qm CF qm D qm E = the ratio of r to i for total investment, = the ratio of r to i for investment out of cash flow, = the ratio of r to i for investment out of new debt, = the ratio of r to i for investment out of new equity. 17
Hypothesis 5: For companies with qm I > 1, it is also true that qm CF > 1, qm D > 1, and qm E > 1. Hypothesis 6: In countries with weak capital market discipline, for companies with qm I < 1, 1 > qm D > qm CF and 1 > qm D > qm E. 18
Methodology I t : a firm's investment in period t, CF t+j : the cash flow this investment generates in t+j, i t : the firm's discount rate in t. The present value of this investment is PV t = j= 1 CF t+ j ( 1+ i ) t (1) 19
Define a permanent return, r t, which creates an equivalent present value to that defined by (1). Ir PV = t t= qm I (2) t t t it where qm t = r t / i. t The market value of the firm at the end of period t can be defined as M = M + PV δ M + µ (3) t t 1 t t t 1 t 20
M M I = δ + qm + M M M µ t t 1 t t t 1 t 1 t 1 (4) I = After tax profits + Decpreciation Dividends + D + E + R&D + ADV 21
The Data The financial data: 1996-2001 versions of the Global Vantage, 1997 version of the Compustat from S & P's. We exclude banks and financial companies and some service industries (SICs 6000 through 6999 and above 8100) 22
To minimize the weight of outliers, we cap our basic variables at both the 1 st and 99 th percentiles of each country sample. After this procedure we are left with 19,010 companies. Ownership data: AMADEUS database for the European companies, Compact Disclosure for the USA, Asian (Japanese) Company Handbook for Asian companies (Japan). 23
Estimated qms by Legal Origin Legal System English Origin Scandinavian Origin German Origin French Origin Transition Countries African Countries 1.02 0.78 0.74 0.59 0.64 0.77 qm t-value 111.34 13.67 35.51 29.23 4.30 10.90 n = 112, 590, Adj.R 2 = 0.25 24
Country Australia 0.94 Bermuda 0.91 Canada 1.16 Cayman Islands 0.58 Great Britain 0.85 Hong Kong 0.78 India 0.80 Ireland 1.10 Estimated qms by Country qm (qm 1) 0.20 0.27 0.00 0.02 0.00 0.01 0.04 0.21 Firms 346 215 1478 42 1331 127 246 63 25
Estimated qms by Country Country Israel Malaysia New Zealand Pakistan Singapore South Africa Thailand USA 1.27 0.86 0.86 0.40 0.97 1.07 0.64 1.05 qm (qm 1) 0.18 0.00 0.05 0.00 0.75 0.72 0.00 0.00 Firms 56 381 66 46 208 118 243 8591 26
Estimated qms by Country Scandinavian Origin Country Denmark Finland Norway Sweden 0.65 0.96 1.04 0.65 qm (qm 1) 0.00 0.69 0.63 0.00 Firms 101 79 103 156 27
Estimated qms by Country German Origin Country Austria Germany Japan South Korea Switzerland Taiwan 0.71 0.57 0.86 0.70 0.64 1.26 qm (qm 1) 0.02 0.00 0.00 0.00 0.00 0.01 Firms 82 425 2219 82 160 126 28
Estimated qms by Country French Origin Country qm (qm 1) Firms Argentina Belgium Brazil Chile Columbia France Greece 0.78 0.51 0.25 1.24 0.43 0.57 0.54 0.16 0.00 0.00 0.29 0.00 0.00 0.00 24 79 133 73 15 495 49 29
Estimated qms by Country Country Indonesia Italy Luxembourg Mexico Netherlands Netherl. Antil. Panama 0.84 0.64 0.70 0.50 0.69 1.19 1.25 qm (qm 1) 0.06 0.00 0.52 0.00 0.00 0.17 0.23 Firms 132 150 12 81 174 19 4 30
Estimated qms by Country Country Peru Philippines Portugal Spain Turkey Venezuela China 0.11 1.00 0.46 0.54 0.54 0.58 0.45 qm (qm 1) 0.00 0.98 0.00 0.00 0.00 0.04 0.00 Firms 20 83 49 117 29 10 70 31
B. The Effects of Ownership Structures Family Financial Non-Financial Dispersed State English Origin 1.082 > 1.019 1.002 < 1.061 1.058 1.041 1.001 1.050 1.014 1.045 Scandinavian Origin 0.773 0.739 0.561 0.812 0.718 0.761 1.145 > 0.683 0.796 0.743 32
B. The Effects of Ownership Structures Family Financial Non-Financial Dispersed State European-German Origin 0.599 0.636 0.561 0.640 0.626 0.628 1.358 > 0.601 0.374 < 0.634 Asian-German Origin 0.977 0.872 0.869 0.882 0.896 0.863 0.829 0.906 1.322 0.880 33
B. The Effects of Ownership Structures Family Financial Non-Financial Dispersed State French Origin 0.569 0.605 0.692 > 0.579 0.565 0.644 0.543 0.605 0.952 > 0.588 34
C. The Effects of Insider Ownership on qm in the United States Insider Ownership = 0 Insider Ownership = 22 % Insider Ownership = 68 % Insider Ownership > 68 % qm = 0.95 qm = 1.21 qm = 0.92 qm rises 35
In Europe: Effects of Pyramiding (P), Cross Shareholdings (C) Voting Rights = Cash Flow Rights (V) qm = 0.80 when P = 0, C = 0, V = 0 qm = 0.34 when P = 1, C = 1, V =1 36
Effects of Business Groups: Japan 1985 2000 1985 1995 1996-2000 Independent 0.94 1.07 0.59 Group 0.83 0.92 0.57 37
qm by Source of Funds English Origin Sample % Median CF Debt Equity All 1.09 0.86 1.09 1.37 qm 1 55 1.74 1.48 1.35 1.99 qm < 1 45 0.51 0.36 0.77 0.63 38
qm by Source of Funds Scandinavian Origin Sample % Median CF Debt Equity All 0.85 1.31 1.08 0.55 qm 1 42 1.56 2.29 1.42 1.37 qm < 1 58 0.55 0.71 0.86 0.21 39
qm by Source of Funds German Origin Sample % Median CF Debt Equity All 0.84 0.70 0.98 1.09 qm 1 39 1.45 1.57 1.27 1.59 qm < 1 61 0.55 0.47 0.83 0.55 40
qm by Source of Funds French Origin Sample % Median CF Debt Equity All 0.78 0.64 1.02 0.52 qm 1 38 1.67 1.39 1.41 1.10 qm < 1 62 0.46 0.46 0.84 0.37 41
qm by Source of Funds Sample No of firms Median CF Debt Equity Transition All 78 0.76 0.39 1.25 1.29 Africa All 17 0.71 0.45 0.90 1.05 China All 48 0.6 0.28 1.14-0.46 42
Effects of Accounting Standards English Origin Sample Standard CF Debt Equity All Weak 0.83 1.05 1.07 All Strong 1.32 1.05 1.86 qm 1 Weak 1.62 1.47 1.74 qm 1 Strong 1.88 1.30 2.39 qm < 1 Weak 0.37 0.78 0.57 qm <1 Strong 0.68 0.70 0.93 43
Effects of Accounting Standards Scandinavian Origin Sample Standard CF Debt Equity All Weak 1.28 1.04 0.64 All Strong 1.39 0.97 1.01 qm 1 Weak 1.47 1.48 0.69 qm 1 Strong 2.38 1.22 2.19 qm < 1 Weak 1.23 0.99 0.56 qm <1 Strong 0.72 0.76 0.47 44
Effects of Accounting Standards German Origin Sample Standard CF Debt Equity All Weak 0.59 0.99 1.03 All Strong 1.09 0.95 1.33 qm 1 Weak 0.81 1.10 1.58 qm 1 Strong 2.01 1.27 1.77 qm < 1 Weak 0.58 0.89 0.66 qm <1 Strong 0.62 0.72 0.72 45
Effects of Accounting Standards French Origin Sample Standard CF Debt Equity All Weak 0.73 1.05 0.65 All Strong 0.89 1.07 0.98 qm 1 Weak 1.49 1.44 0.84 qm 1 Strong 1.48 1.44 1.45 qm < 1 Weak 0.51 0.88 0.50 qm < 1 Strong 0.58 0.79 0.57 46
Effects of Creditors Rights English Origin Sample Standard CF Debt Equity All Weak 1.32 1.05 2.06 All Strong 1.34 1.03 1.06 qm 1 Weak 1.87 1.31 2.54 qm 1 Strong 2.07 1.34 1.57 qm < 1 Weak 0.65 0.70 1.06 qm < 1 Strong 0.78 0.71 0.58 47
Effects of Creditors Rights Scandinavian Origin Sample Standard CF Debt Equity All Weak 1.40 0.97 1.01 All Strong 1.29 1.04 0.64 qm 1 Weak 2.37 1.21 2.19 qm 1 Strong 1.46 1.47 0.69 qm < 1 Weak 0.72 0.76 0.48 qm <1 Strong 1.22 0.99 0.57 48
Effects of Creditors Rights German Origin Sample Standard CF Debt Equity All Weak 1.09 0.94 1.33 All Strong 0.59 0.98 1.03 qm 1 Weak 2.01 1.26 1.77 qm 1 Strong 0.81 1.10 1.58 qm < 1 Weak 0.62 0.72 0.72 qm <1 Strong 0.58 0.89 0.66 49
Effects of Creditors Rights French Origin Sample Standard CF Debt Equity All Weak 0.78 1.04 0.78 All Strong 0.85 1.18 0.42 qm 1 Weak 1.45 1.41 1.08 qm 1 Strong 2.04 1.83 0.30 qm < 1 Weak 0.55 0.85 0.54 qm <1 Strong 0.41 0.84 0.36 50
Robustness Check Possible problems 1. Efficient capital market 2. Investment endogenous 51
M t = M t-1 + q mt I t - δ t M t-1 + µ t (6) M t+1 = M t + q mt+1 I t+1 - δ t+1 M t + µ t+1 (7) M t+1 = M t-1 + q mt I t + q mt+1 I t+1 - δ t M t-1 - δ t+1 M t + (8) + µ t + µ t+1 52
n n n M = M + q I M + δ µ t+ n t 1 mt+ i t+ i t+ i t 1+ i t+ i i= 0 i= 0 i= 0 (9) q m n q I mt+ i t+ i = i= 0 n It+ i i= 0 (10) 53
q m n i = 0 q n i = 0 I mt+ i t+ i = = I t+ i M M = + n δ µ M t+ i t 1+ i t+ i t+ n t 1 i= 0 i= 0 n n n I I I t+ i t+ i t+ i i= 0 i= 0 i= 0 n (11) 54
q m Compute s for firms with 3 or more observations qˆ 1.03 q, m = 2 m R = 0.98 (weight by number of firms in country). 55
Table 8 Comparison of Estimated ( q ) ˆm and Calculated ( ) Returns on Investment q m The table reproduces the estimates of returns on investment by legal system from Table 2 ( qˆm ) and compares them with means of calculated returns on investment using eq. 11 ( ). q m q m Legal System English Origin qˆ m 1.02 Mean 1.04 Weighted Mean 1.01 Scandinavian Origin 0.78 0.87 0.74 German Origin 0.74 0.83 0.66 French Origin 0.59 0.58 0.62 56
Conclusions Corporate governance institutions and ownership structures are important in explaining differences in company returns on investment relative to their costs of capital, qm. 57
(1) The origin of a country s legal system is the most important. English-origin legal systems provide shareholders with more protection against managers than other systems. (2) Differences in investment performance related to a country s legal system dominate differences related to ownership structure. (3) A hierarchy of finance with respect to the returns on investment 58
Accounting standards and rules that protect creditors rights have significant effects on the returns on investment from new equity and new debt issues, even in countries with generally strong corporate governance systems. Agency problems exist in all countries and can have significant impacts on the investment performance of companies. 59
Agency problems can be mitigated by the institutional structures of a country. Legal institutions that strengthen shareholder rights do bring about superior investment performance. Strengthening external capital market constraints can also increase returns on investments out of externally raised finance. 60