Private Profit or Public Purpose? Shallow Convergence on the Shareholder Model Abstract: Evidence from twelve countries between 1989 and 1999 with an equity capitalization of $10.6 trillion suggests “shallow convergence” on the Anglo-American model of minority investor protections. There has been widespread reform of information and management practices but minimal adoption independent oversight or contests for control. Private blockholders and employee-managers have done little to support these minority investor protections, however, and have resisted these changes in many cases. Instead, process tracing of regulatory changes in these countries points to the state as the prime mover of governance reforms, motivated by the privatization of state-owned enterprises, moral hazard losses from poor governance, and pension funding costs. What is causing corporate governance practices around the world to provide greater minority investor protections – private markets or public purpose? A sample of twelve European and Asian countries between 1989 and 1999 provides a partial answer to this question, with a surprising twist.1 States rather than private majority shareholders appear to be the engine of change in corporate governance practices worldwide. Foreign portfolio investors, largely based in New York and London, doubled their share of total shareholdings in these countries over the past decade, while demanding improved minority shareholder protections. On the supply side, over that same time period, these countries adopted governance practices that provide enhanced minority investor protections, often referred to for convenience as the “Anglo-American” governance model. These practices include considerable reform in information and management incentive practices, and limited adoption of independent oversight or control institutions. The pattern of conformance with the Anglo-American model varies widely by country and among institutions, but the direction of change is unidirectional, towards enhanced minority protections. Is the relationship between this wave of foreign portfolio investment and corporate governance reforms a straightforward process of supply and demand, reaching a new equilibrium of minority protections in response to share price incentives? Or is something more complex going on? Above all, what explains 1 This sample includes six European (Belgium, France, Germany, Italy, the Netherlands, and Spain) and six Asian countries (China, Japan, Korea, Malaysia, Singapore, and Taiwan), which together account for 42% of the MSCI Global Market Index weighted by GDP, 27% of the total MSCI market capitalization in current dollars, and over 75% of the non-U.S. and non-U.K. MSCI market value. the variation in response to these convergence pressures among the countries in this sample? Two Models of Change Two contrasting models of institutional change can be extracted from the literature on international governance change. These models are rarely explicitly invoked, and are usually obscured by the spirited debate whether governance changes constitute full convergence on the Anglo-American model, shallow convergence, functional convergence, or idiosyncratic country-level “path dependence”. The first is the private markets model, and the other I term the public purpose model. In the private markets model, familiar to economists and widely assumed by most financial professionals, the chain of causation that leads to changes in governance practices begins with foreign portfolio investment (FPI), on the demand side. These global investors pay a premium for shares in firms with good governance. This price premium provides private blockholders with an incentive to supply enhanced minority protections in exchange for higher valuation by foreign investors, adopting better governance practices at the level of the firm while lobbying their governments for regulatory reform. The public purpose model starts the same way, with the FPI premium for improved minority shareholder protections. But the state is the prime mover in governance changes, rather than private blockholders. The government adopts formal changes in governance practices because it is primarily motivated by the prospect of proceeds from privatizing state-owned enterprises (SOE’s), moral hazard losses, and pension funding problems. These enhanced minority shareholder protections impose a “tax on blockholders” by making it harder for them to extract private benefits of control, thereby inducing a gradual process of reluctant governance reforms on the part of private blockholders. In this model, the chain of causation between demand for enhanced minority protections and the supply of reforms is more variegated, follows several pathways, and initially has a public rather than private actor on the supply side. Each of these models invokes different theoretical arguments and empirical evidence from the corporate governance literature in specifying the causal link between financial globalization and governance changes, although they both begin with FPI and result in changes in corporate governance. In the next section I briefly summarize the evidence linking FPI and changes in corporate governance practices. I then examine the theoretical underpinnings of each causal model, comparing the predictions of the private markets and public purpose models with evidence from the sample. The paper ends with a brief discussion of the theoretical and public policy implications of these findings. The Wellspring of Good Governance Both causal models share the premise that Anglo-American investors control the bulk of global equity flows, carrying their preferences for minority investor protections with them when they go abroad. Table 1 shows that these United States (U.S.) and United Kingdom (U.K.) investors account for three-quarters of the financial asset pool and 87% of all equity holdings among the “Big Five” economies. Table 1 : Institutional Investor Size, 1999 ($ billions)2 Country Total Assets % in Equity Share of Total Equity U.S. $15,800 .45 .72 U.K. $2,200 .67 .15 Japan $3,200 .19 .06 France $1,200 .30 .04 Germany $1,200 .19 .02 It is also clear that the foreign ownership as a percentage of market capitalization in the twelve countries in question almost doubled from 11.2 in 1989 to 21.2 in 1999. Table 2 shows the growth in FPI ownership of the total value of each country’s market capitalization over the decade period. As a result, money- 2 Carolyn Brancato, “International Patterns of Institutional Investment,” The Conference Board Institutional Investment Report, April 2000, page 9. managers in New York and London became minority shareholders with, on average, a fifth of the control rights over all traded firms in these sample countries, in a relatively short period of time. Table 2: Changes in Foreign Equity Ownership 50 45 Belgium 40 China 35 France Germany 30 Italy FPI (%) 25 Japan Korea 20 Malaysia 15 Netherlands Singapore 10 Spain Taiwan 5 0 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Year The value of stock markets in these countries soared during this period of FPI influx. The market capitalization of traded equities in these countries trebled in percentage terms from 1989-99 – a stunning increase in wealth, from $4.4 trillion to $10.6 trillion (in current dollar equivalents). Minority Protections and the Good Governance Premium However, the governance literature provides conflicting conclusions about how minority shareholder protections affect firms’ financial performance by reducing either agency or expropriation costs, and whether these institutional protections are reflected in higher or lower share prices.3 Most of the empirical work so far uses data from the United States, where there is increasing support for the proposition that minority investor protections do improve firms’ financial performance and, by extension, share prices.4 For example, a recent study by Gompers et.al. found a significant correlation between 3 This literature is summarized in several surveys including Andrei Schleifer and R. Vishny, 1997, “A Survey of Corporate Governance”, Journal of Finance, 52, 737-83; 4 Ira Millstein and MacAvoy, 1998, “The active board of directors and performance of the large publicly traded corporation,” Columbia Law Review, 98. minority investor protections and stock market valuation, based on an index including 24 oversight, control, and management institutions for 1500 firms between 1990 and 1999. Firms in the lower decile of the index (stronger protections) earned abnormal returns of 8.5% higher than those in the upper decile (weaker protections).5 Foreign portfolio investors also appear to factor these institutional practices into share prices abroad, based on statistical, survey, and anecdotal evidence. 6 For example, a McKinsey survey of two hundred investors found the average good governance premium for a sample of twenty-two countries to be 21.6%, with a low of 17.9% and a high of 27.6%; for the ten countries used in this paper’s sample, the average premium was 20.9%.7 Mitton analyzed a sample of four hundred firms in Korea, Malaysia, the Philippines and Thailand before and after the Asian Financial Crisis of 1997-98 and found statistical evidence for a 12% premium valuation on firms that had adopted superior information institutions; moreover, firms that followed these two information practices came through the perturbations of the Asian Financial Crisis trading at a 58% premium over those Asian firms that did not follow these practices.8 However, there remain formidable obstacles to using the tools of statistical inference to establish a causal link between the governance practices of individual countries and the value of their firms as measured by either abnormal share returns or some other proxy for valuation, such as Tobin’s Q. Much of the reported financial data from firms is spotty and use different accounting practices. So many independent variables affect country stock market valuation that the effects of corporate governance practices are easily lost in the “noise”. Moreover, there is only a small number of countries with reliable data, which therefore limits the power of statistical tools to extract the impact of corporate governance practices on share prices. As a result, both the private markets and public purpose models stand on shaky empirical ground on this point, with much econometric work still to be done. 5 Paul Gompers, Joy Ishii, Andrew Merrick, “Corporate Governance and Equity Prices”, NBER Working Paper 8449, August 2001, www.nber.org/papers/w8449. The authors make no claims about the arrow of causation between good governance and performance, however. 6 Stephen Davis, “Leading Corporate Governance Indicators: An International Comparison,” November 1999, www.dga.com; Sang Yong Park, “Value of Governance of Korean Companies: International Investors Survey,” manuscript, April 1999. 7 Robert Felton, Alec Hudnut, and Jennifer van Heeckeren, “Putting a value on Board Governance,” The McKinsey Quarterly, No. 4 (1996); McKinsey Investor Opinion Survey, June 2000, www.gcgf.org. In this survey, “good governance” was defined as a majority of outside directors, extensive disclosure, and use of stock options for directors – a rather narrow range of governance institutions. 8 Todd Mitton, “A Cross-Firm Analysis of the Impact of Corporate Governance on the East Asian Financial Crisis,” MIT manuscript, 3/15/2000. In these tests of the effects of information institutions, Mitton controlled for firm size and financial leverage, and included both country and industry dummy variables, with a sample size of 384 firms based on Worldscope data. The Shifting Terrain of Governance In the face of this presumed “good governance premium” offered by foreign portfolio investors, the pattern of conformance in governance practices is remarkably varied, although Table 5: The Shifting Terrain of Governance the direction of change is INFORMATION Accounting Audit OVERSIGHT NED Duty CONTROL Voting Takeover MANAGEMENT TOTAL Incentive consistently towards the Belgium China 2 0 Anglo-American model. The France Germany 3 1 three charts in Table 5 Italy Japan 0 1 operationalize seven corporate Korea Malaysia 0 3 governance institutions in Netherlands 1 Singapore 4 terms of their conformance to Spain 0 Taiwan 0 the Anglo-American TOTAL 0 5 3 2 4 0 1 15 governance model in a series of GROUP TOTAL % Conform 0.0% 41.7% 25.0% 16.7% 33.3% 0.0% 8.3% 17.9% 5 5 4 1 “snapshots” taken in 1989, % Conform 20.8% 20.8% 16.7% 8.3% 1994, and 1999.9 Shaded boxes Corporate Governance Conformance: 1994 INFORMATION OVERSIGHT CONTROL MANAGEMENT TOTAL indicate protection for Belgium Accounting Audit NED Duty Voting Takeover Incentive 2 minority investors in line with China 1 France 4 the expectations of Anglo- Germany 1 Italy 0 American investors, and white Japan 1 Korea 0 Malaysia 5 Netherlands 3 9 Singapore 5 Accounting systems are coded 1 Spain 1 (black box) if a majority of the listed Taiwan 1 TOTAL 4 8 3 2 4 0 3 24 firms use GAAP or IAS in their % Conform 33.3% 66.7% 25.0% 16.7% 33.3% 0.0% 25.0% 28.6% reporting, or if the deviations of the GROUP TOTAL 12 5 4 3 country’s domestic accounting % Conform 45.8% 20.8% 16.7% 25.0% Corporate Governance Conformance: 1999 systems from IAS are minor. The INFORMATION OVERSIGHT CONTROL MANAGEMENT TOTAL former datum is taken from the Accounting Audit NED Duty Voting Takeover Incentive Belgium 2 reports of country stock exchanges China 1 or survey agencies; the latter is based France 5 on the opinions of professional Big Germany 3 Italy 3 Five auditors interviewed by the Japan 2 author. Audit is coded 1 if third-party Korea 5 Malaysia 6 audit is a listing requirement. Netherlands 3 Oversight is assigned using the mean Singapore 6 Spain 3 percentage of non-executive directors Taiwan 3 on the boards of listed firms; where TOTAL 8 10 5 6 6 0 7 42 NED’s are a majority, oversight is % Conform GROUP TOTAL 66.7% 83.3% 18 41.7% 50.0% 11 50.0% 0.0% 6 58.3% 7 50.0% coded 1 (except for Germany and the % Conform 75.0% 45.8% 25.0% 58.3% Netherlands). Fiduciary duty is coded 1 if directors’ liability to minority shareholders has been enforced in the courts on the basis of derivative or class-action suits. Voting rights is coded 1 if the principle of “one share one vote” is observed in practice, in terms of statutory rights and procedures. If more than 50% of the listed firms employ significant anti-takeover provisions, the box is coded 0. Incentive is coded 1 if the sum of performance bonus and stock options exceeds 10% of total pay, based on a Towers & Perrin survey. These conformance rankings are consistent with those of Davis Global Advisors and similar to those performed by Déminor, with a few exceptions. For a discussion of the pitfalls of governance rankings, see Eric Coppieters, “The Governance Scorecard” and “Governance Ratings in Europe”, Corporate Governance International, March 2001. boxes indicate little or no such protection. These rankings are based upon the best data and indices available, although prone to error and an element of subjective judgement.10 As summarized in Table 6, on a country basis, the average conformance level for this sample almost trebled over the decade, from 18% to 50%. By1999, the sample ranged from a low of 1 (China) to a high of 6 (Singapore). The average conformance for the Asian countries as a group was slightly higher than those of the European countries in the sample. Two-thirds of the changes took place in the five years between 1994-99. Table 6: Conformance Level Summary Conformance Levels Changes in Conformance Levels 1989 1994 1999 89-94 94-99 89-99 Belgium 2 2 2 0 0 0 China 0 1 1 1 0 1 France 3 4 5 1 1 2 Germany 1 1 3 0 2 2 Italy 0 0 3 0 3 3 Japan 1 1 2 0 1 1 Korea 0 0 5 0 5 5 Malaysia 3 5 6 2 1 3 Netherlands 1 3 3 2 0 2 Singapore 4 5 6 1 1 2 Spain 0 1 3 1 2 3 Taiwan 0 1 3 1 2 3 Total 15 24 42 9 18 27 Mean 1.25 2.00 3.50 Avg % 0.18 0.29 0.50 Europe % 0.17 0.26 0.45 Asia % 0.19 0.31 0.55 Eur changes 4.00 8.00 12.00 Asia changes 5.00 10.00 15.00 Looking at institutional practices rather than countries, conformance levels for accounting and audit practices increased from 20% to 75%, in both Asia and Europe. Conformance oversight doubled, from 21% to 46%. There was little change in voting rule conformance, from 33% to 50%, and zero change in takeover rules, reflecting a blanket refusal to adopt the institution of market- 10 The author tracked down the causes of changes in governance institutions in each country based on 120 field interviews with investors, financiers, private blockholders, managers, and government officials during 2000-2001. As a professional manager and then as an outside director, the author incorporated or managed a firm in 9 of the 12 countries in the sample, and therefore had access to a network of bankers, attorneys, accountants, and search firms with whom to verify both institutional practices and the proximate source for these changes. based contests for control, a.k.a. hostile takeovers, in both the Asian and European sample. The practice of using long-term incentives (principally stock options) for managers increased dramatically, from 8% to 58%. In sum, there is reasonably good empirical support for the argument that foreign portfolios pay a price premium for shares in firms and in countries with better minority shareholder protections. The weight of FPI has increased substantially in this sample, and the pattern of conformance to the Anglo-American model apparently reflects this price incentive at work, although with considerable variation between countries and among institutions. What drove this process of change– private markets or public purpose? The private markets and public purpose model provide different explanations for what links the FPI governance premium and the pattern of changes in these conformance levels. The Private Markets Model According to the private markets model, the prime movers on the supply side of governance reform are the majority shareholders (“private blockholders” in governance parlance) who control most publicly-traded firms abroad. Prior to the influx of FPI, private blockholders and employees supported domestic governance practices that allow the former to extract expropriation costs and the latter to extract agency costs from the firm, both at the expense of minority shareholders.11 Face with the potential good governance premium from FPI, private blockholders defect from this arrangement and trade their private benefits of control for higher valuation by FPIs. Private blockholders encourage professional managers support these new practices (and abandon any sense of solidarity with line employees) by means of stock options that align them with the shareholders rather than other employees. Blockholders then ensure that firms adopt informal best practices where they can, such as enhanced disclosure and independent oversight, while lobbying the state for formal regulatory where governance practices are fixed by statute. The state is a passive agent of voters and special interests (such as private blockholders). The agent state responds accordingly to lobbying pressures by these private principals: formal governance regulations gradually change to provide enhanced minority investor protection. Private blockholders gradually sell down their concentrated holdings to portfolio investors, trading their private benefits of control for higher valuation. 11 See Marco Pagano and Paolo Volpin, “The Political Economy of Corporate Governance,” paper presented at the European Corporate Governance Network, 10/99. Predictions and Evidence How well does the private markets model hold up to the evidence from this twelve-country sample? Is there evidence of private blockholders accepting a “good governance” deal from FPIs? Recent studies on underlying ownership suggest that private blockholders do control a large percentage of listed firms (weighted by market value) in most countries in this sample. Table 7: Ownership Pattern (%)12 Country Private Blockholder State Employee-Manager Belgium 47 13 40 China 0 100 0 France 50 20 30 Germany 40 20 40 Italy 80 10 10 Japan 8 2 90 Korea 68 5 27 Netherlands 20 10 70 Malaysia 60 25 15 Spain 30 30 40 Singapore 60 30 10 Taiwan 65 5 30 Mean 44 23 33 2 25 26 25 Compared to the mean value of the sample, three countries are clearly dominated by private blockholders: Italy, Korea, and Taiwan. Many of the Korean corporate groups, or chaebol, show small nominal ownership percentages by the controlling family; but when the veil of indirect control is pierced, the underlying pattern of concentrated family control is revealed – over two-thirds of the market value is in the hands of a few families. Singapore and Malaysia also have high private blockholder concentrations, mostly in the hands of a small 12 European cases are based on data collected by Christoph van der Elst, “The Equity Markets, Ownership Structures and Control: Towards International Harmonization?” Universiteit Gent Working Paper 2000-04, 9/2000; European Corporate Governance Network, www.ecgn.org; Marco Becht and Colin Mayer, The Control of Corporate Europe, forthcoming; and Becht and Boehmer, op. cit. The Asian cases are based on data collected by Claessens op. cit. plus papers from the OECD Forum on Corporate Governance in Asia and the author’s own calculations. number of Overseas Chinese (hua qiao) families, coexisting with a large state- controlled sector. Belgium, France, and Germany are above the mean for both private blockholders and manager-controlled firms, with a relatively small state-owned sector. Spain is an ambiguous case, close to the mean for all three ownership types. In China the state dominates as the ownership type, while Japan and the Netherlands are dominated by manager-controlled firms. In Japanese manager-controlled firms there is ultimately no private shareholder, no blockholder, with a motive or means to maximize shareholder value.13 This suggests that for at least 3 and possibly 4 country cases in the sample, private blockholders are too few or simply absent as a engine for change in corporate governance practices, as assumed by the private markets model. For the 8 countries that do have a weight of private blockholders above 40%, the private profit model predicts that private blockholders will unilaterally adopt good governance standards for informal institutions they can control at the firm level and then lobby their governments for changes in formal institutions. This is not borne out by evidence from sample. Table 8 ranks governance institutions by increasing formality. Firms are free to innovate in the first three areas; the latter four require formal changes in government statutes. The rate of change is equal in both groups of practices, with no discernable pattern of informal change preceding formal changes. Table 8: Informal and Formal Change (units, from table 4) Institution 1999 Conformance 1989-99 Change Informal Practices (set by management norm) Management: incentive compensation 5 4 Oversight: outside directors 5 2 Control: anti-takeover devices 0 0 Formal Practices (set by government statute) Information: audit 7 3 Information: accounting systems 7 7 Control: voting 5 2 Oversight: fiduciary responsibility 5 3 13 “Cross-shareholdings have altered the concept of the stock company in Japan. They have freed management substantially from the influence of shareholders. Stable shareholders are selected by and are dependent on management, and cannot therefore play a disciplinary role on corporate management…..This leaves the main source of disciplinary pressure coming from the need to satisfy employees.” Seiichi Masuyama, “The Role of Japanese Capital Markets” in N. Dimsdale and M. Prevezer, editors, Capital Markets and Corporate Governance, page 334. Digging deeper into the country cases, there is little evidence that private blockholders lobbied the state for governance changes, either singly on a collective basis. For every country in this sample in which business firms engage in collective lobbying through a group such as the Federation of Korean Industry (FKI), Germany’s Bundesverband der Deutscher Industrie (BDI) or Spain’s Circulo de Empresarios, these collective business entities have been passive or hostile with regard to corporate governance reforms, despite active lobbying across a wide range of other regulatory issues. Business federations have been most hostile to governance reforms in the three countries with the highest concentration of private blockholders: Italy (80%), Korea (68%), and Taiwan (65%). As for the private market model’s prediction that private blockholders should incrementally liquidate their concentrated holdings to take advantage of the FPI premium for good governance, there is no evidence of this taking place in any country in the sample. Although confirmation of changes in the portfolio concentration of private blockholders must wait until a second round of control studies in Europe and Asia provides a second set of data points, scattered and anecdotal evidence for this twelve country sample suggests no liquidation of control by private blockholders. In Korea, the top thirty chaebol blockholders have increased their concentration within affiliated groups in recent years, in the face of pressure from the state and incentives from FPI to do so. 14 Anecdotal evidence from Italy, Taiwan, Malaysia, and Singapore shows no signs of large private blockholders reducing their holdings in order to profit from higher FPI valuation; on the contrary, as in Korea, many owner-families have been concentrating their ownership holdings. Nor have private blockholders of established firms in any of the European countries in this sample engaged in a sell-down to FPI’s - though some have sold out to multinational corporations or to private equity funds.15 No-go on the Good Governance Deal What explains this apparent failure of private blockholders to accept the “good governance deal” from FPI’s? Either blockholders were not motivated to adopt governance reforms because the private benefits continued to outweigh the FPI valuation benefits, or the blockholders were motivated to accept the “good governance deal” but were unable to press through governance reforms for other reasons. 14 Myeong-Hyeon Cho, “Corporate Governance in Korea”, paper for UCSD Conference on Corporate Restructuring in Korea, 10/2000. 15 The ownership pattern data presented in Table is static - essentially a snap-shot, drawn for this sample from the first wave of ownership disclosure in Europe and Asia that began in the mid-1990’s. Until this analysis is repeated to provide at least a second set of more recent data points, it will not be clear how much ownership concentration has changed overall, and thus whether liquidation has taken place. It is possible that private benefits are extremely hard to replace with higher market valuation, for a combination of tax, psychological, and uncertainty. The good governance deal exposes the private blockholder to the tax collector, whereas private benefits can be hidden or routed through offshore tax havens. Anecdotal evidence suggests that many of the countries with the highest private blockholder weights also exhibit patterns of capricious taxation by the authorities and widespread tax evasion by business owners, as in Italy, Korea, Malaysia, Spain, and Taiwan. Stock market valuation can motivate blockholders with money, but it cannot compensate blockholders for the psychological benefits of control, in terms of prestige, nationalism (as in the case of the chaebol founder-chairmen whose firms built the Korean “economic miracle”), or the satisfaction of placing children into jobs in the family firm. The good governance deal involves a good deal of uncertainty. Stock markets themselves can be capricious, and blockholders observed the rapid run-up and abrupt crash of emerging market valuations during the latter half of the 1990’s, which increased the discount rate they assigned to the “good governance deal”. In contrast to these market uncertainties, although the private blockholder is exposed to high firm risk because of his concentrated holdings, he also has the benefit of insider knowledge and (usually) accumulated expertise in his industrial sector. Moreover, since many blockholders control a horizontally diversified group of firms – with some family firms in Asia holding horizontal interests in industries including flour milling, semiconductors, and banks – the portfolio gains from the good governance deal may be limited. If on the other hand these problems can be circumvented and private blockholders are still motivated to accept the “good governance deal”, as predicted by the private markets model, they still face a formidable transactions problem in selling off ownership in small increments to foreign portfolio investors while recovering the control premium. For a smooth transition to take place, the blockholder must issue a credible promise not to steal from the firm, and the FPI’s must promise to compensate the blockholder for giving up those private benefits of control – in effect, paying a premium over the current value of traded minority shares in that firm. If done in small increments, the marginal transaction that transfers control from the blockholder to the new minority investor must carry a price that embodies all the private benefits of control – a big price tag. No rational minority investor will enter this transaction, since all the previous shares were acquired at the lower traded price. Yet without such a price premium, no rational blockholder will sell to the point where control is threatened. Unless this transaction problem is solved, blockholders’ support for the “good governance deal” may grind to a halt at the transfer of the 51st percent of control, leaving the private blockholder in charge and minority shareholders still exposed. Private Blockholders’ Collective Action Hurdle Even if blockholders are motivated to accept the governance deal and find a way around the control transactions problem, they encounter two collective action problems that may explain their apparent inaction in supporting governance reforms. The first problem is trust. In a market dominated by other blockholders and with weak governance institutions to protect minority shareholders, what blockholder will be brave enough to go first, adopting good governance institutions that limit his own private benefits, and exposing himself to expropriation as a trusting minority shareholder? If the solution to “who goes first” is collective lobbying for governance reform through the industrial federation, which binds all blockholders to the same new rules, the private blockholders must contend for control of the business federation with the professional managers of semi-privatized state-owned enterprises (SOE’s) and large manager-controlled firms. These are usually larger than private blockholder firms and more skilled at interfacing with the bureaucracy, and they tend to dominate the agenda of big business federations in many economies, exhibiting only limited sympathy for the efforts of private blockholders to change the governance rules to enrich themselves. For example, in France, blockholders engaged in an extended struggle for control of the corporate governance agenda of the Mouvement des Éntreprises de France (MEDEF) with professional employee-managers, especially the cadre of state managers known as “ENArchs”, a term for graduates of the elite schools such as the École Nationale d’Administration (ENA) who move back and forth between government posts and managerial positions in the state sector. German blockholders also failed to mobilize the Bundesverband der deutscher Industrie (BDI) in the pursuit of corporate governance reforms, in the face of resistance by the Big Three banks and a handful of large firms controlled by managers. On balance, the evidence from this twelve country sample does not support the predictions of the private markets model, and most of the anecdotes suggest that the “good governance deal” between FPI and private blockholders never made it past the first stage, blockholders apparently deciding that the private benefits of control continue to outweigh the gains from higher valuation. If the evidence does not support the private markets model at work, then how are we to account for the remarkable shifts in the governance terrain in this twelve country sample? The Public Purpose Model According to the public purpose model, the changes on the supply side of corporate governance are driven by the state – an active state, not a passive “agent” state, and a state with vital stakes in corporate governance practices. From the standpoint of governments – defined as the bureaucratic elites who calculate budgets and the politicians who must submit these budgets (and taxes) to voters – corporate governance reforms in the direction of the Anglo-American model provide a partial solution to the fiscal squeeze faced by all the countries in this sample over the last decade. The state’s motives for supporting the Anglo- American model include income from the privatization of previously state- owned enterprises, moral hazard losses from poor governance (especially in the financial sector), and the funding of state pension plans. The state was transformed from being a passive agent by the fiscal costs that poor governance imposed on national budgets – a hard budget constraint. The data presented in Table 6 demonstrate that the state remains an important controlling owner of listed companies, a public blockholder, in ten out of twelve countries in the sample, even after fifteen years of privatization. Public Blockholders’ Incentive Problems It is important to point out that the “public purpose” that drives governance reforms is not some altruistic sense of good government, but rather the urgent desire of politicians to be re-elected and the equally intense affection of bureaucrats for keeping their jobs. The state as blockholder exhibits a peculiar owner’s incentive problem. Though ownership of a state-owned industrial enterprise or bank is nominally held in the name of “the people,” as a practical matter SOE’s are held by bureaucratic agencies, often multiple agencies, under the ultimate control of politicians. Politicians have many objectives in addition to maximizing the value of the state’s blockholding of an SOE, and thus the state’s response to the FPI good governance deal reflects a mix of personal and public policy motives. SOE’s provide politicians with patronage, excess employment (for which they can take credit), and political contributions or outright bribes. These are the equivalent of the private benefits of control, against which politicians must weigh the appeal of the FPI good governance deal. Predictions and Evidence The public purpose model predicts that the rate of change in governance institutions should be proportional to the intensity of the fiscal squeeze on the government in each country, and that reforms which enhance minority investor protections should follow rather than precede periods of fiscal adversity. These governance reforms should be pioneered in the state-controlled sector, either by SOE’s or among state-controlled banking institutions, rather than in the fully private sector, and process-tracing of individual reforms should find the state driving through these reforms in the face of opposition from private blockholders and employee-managers alike. Every country in this sample witnessed a program of corporatisation and partial privatization of SOE’s, a program that began in the 1980’s and picked up speed during the 1990’s. Partial privatization (perhaps “corporatisation” is a more accurate description) was an attractive way out of this squeeze in every country, not only because it reduced the drain on fiscal budgets that came from subsidizing poorly-performing SOE’s, but also because it provided a fast way for states to raise cash without raising taxes. For every country, the privatization process preceded governance reforms pushed through by the state. Table 9 compares the cumulative value of SOE share initial public offerings (SIPO’s) for the twelve country sample between 1989 and 1999 to cumulative government expenditure and fiscal deficits over the same period. Table 9: Revenues from SOE Share Offerings16 SIPO's IN STATE FINANCE, 1989-1999 SIPO SIPO/Exp SIPO/Deficit Country ($mn) (%) (%) Belgium 1,100 0.11 1.2 China 9,494 0.95 10.7 France 41,999 0.78 7.9 Germany 15,170 0.23 4.4 Italy 59,457 1.51 6.4 Japan 42,852 0.71 6.4 Korea 7,250 1.08 15.6 Malaysia 3,773 2.09 57.5 Netherlands 9,365 0.69 11 Singapore 2,568 1.97 n/a Spain 37,085 2.94 16.3 Taiwan 2,574 0.81 n/a 16 Fiscal data from IMF Government Finance Statistic Yearbook, 2000; SIPO revenues from William Megginson, Appendix Detailing Public Share Offerings 1961-2000, http://faculty- staff.ou.edu/M/William.L.Megginson-1, author’s calculations. The states in this sample obtained a total of $231 billion in cumulative SIPO proceeds between 1989 and 1999. As a percentage of cumulative central government expenditure over this same period, SIPO revenue averaged a modest 1.2%, with a high of 2.94% (Spain) and a low of 0.11% (Belgium). SIPO revenues were a crucial “deficit reduction” measure, plugging on average 14% of the fiscal deficit for those governments who operated in the red during this period. Moreover, these data understate the importance of SIPO revenues to the state, since these SIPO’s freed the state from supplying capital funds to these corporatised SOE’s. After the state liquidated part of its holdings, these firms obtained additional capital themselves from secondary equity offerings, bond issues, and loan syndication’s, thereby reducing the cash-flow drain on the state budget and the state’s contingent liability as sovereign guarantor of their debts. As a result, the state in every country in this sample engaged in a partial sell-off of its blockholdings through SOE SIPO’s, a substantial block of which was acquired by FPI’s. In parallel, the state implemented a series of top-down governance reforms that enhanced minority shareholder protections to attract investors to these SIPO’s and maximize the state’s revenue stream from these offerings. Change in conformance levels has a .45 correlation with the ratio of SOE SIPO’s to state expenditure, as seen in the scatterplot below. 2.94 Spain Malaysia Singapore SOE Proceeds/State Expenditure Italy Korea China France Taiwan Japan Netherlands Germany .11 Belgium 0 5 Change in Governance For example, Spain began an aggressive program of partial privatization of SOE’s in the period 1989-99, to offset yawning deficits during the 1990’s and in the face of the 3% deficit caps imposed by the Maastricht Agreement on the Euro. The Spanish state raised $37 billion – 3% of the total government budget, plugging 16% of the fiscal gap - from share offerings of state firms, particularly utilities such as Telefónica, GESA (Gas y Electricidad de España), Endesa, Repsol, and Tabaclera, as well as the state-controlled banks. The Ministry of Finance imposed tougher new audit requirements for listed firms and delegated accounting standards setting to the quasi-independent Instituto de Contabilidad y Auditoria de Cuentas (ICAC). With state encouragement, SOE’s pioneered the use of GAAP or International Accounting Standards (IAS) and Big Five auditors in these SIPO’s, while pioneering the use of options for managers of privatized SOE’s, first widely-used by Telefónica. In Malaysia the state raised 2% of its revenue and plugged 58% of its budget gap with SIPO’s totaling $3.7 billion over the same period. Malaysia’s SIPO’s began with the sale of a 30% share in Malaysian Airlines in 1985, followed by larger transactions in Telekom Malaysia in 1990 and 1992, Tenaga National in 1992, Petronas in 1995, and Malaysia Airports in 1999. In parallel with these offerings, the state imposed a series of reforms in information, oversight, and management practices that gave Malaysia one of the highest conformance rankings in the twelve country sample.17 In Italy, governance reforms were also imposed by the state, particularly the Ministry of Finance and the Commissione Nazionale per le società e la Borsa (CONSOB).18 The Italian government was under intense pressure to meet the Maastricht guidelines for fiscal deficit caps during the 1990’s, and Italy’s pattern of SIPO’s corresponds closely to the state’s program of plugging the budget gap with SIPO revenues, raising $59 billion between 1989-99, the largest amount in the sample, equal to 1.5% of the entire central government budget for that decade. 17 The Companies Act mandates third-party audit, a requirement backstopped by the Kuala Lumpur Stock Exchange (KLSE). Ninety percent of listed companies have at least two NED’s, and the Malaysian Code of Corporate Governance has set a minimum of 30% independent NEDs on boards. Malaysia’s legal system imposes strong standards of fiduciary duty to minority shareholders, and the courts entertain derivative suits for breach of this duty, although class-action suits are not possible. The obligations of directors are monitored by the Government Minority Shareholder Watchdog Committee, which along with the Securities Commission and the KLSE enforces the one-share-one-vote rule. 18 The Ministry of Finance and a bloc of reform politicians altered the legal foundations of Italian publicly- traded corporations, or società per azioni, invigorated the Commissione Nazionale per le società e la Borsa (CONSOB), and privatized the stock exchanges. As a result, information institutions have gradually come into conformance with international standards of accounting systems, standards setting, and audit. The 1998 changes reduced the ownership threshold by which minority shareholders can bring derivative suits against the management or blockholders of a publicly traded firm. The Chinese government began its strategy of corporatisation of SOE’s to raise revenue for the central government in Beijing, which was perpetually tax- starved, and to shift the continuing losses of the SOE’s out of the state-owned banking system by imposing a hard budget constraint, as SOE losses were estimated at 4-5% of GDP. As a result of a string of SIPO’s during this decade, many SOE’s were removed from the central government’s books and brought in revenues equal to 1% of the central budget – thereby plugging 10% of the budget gap. In order to promote the growth of domestic stock markets and to tap international equity investors, particularly in Hong Kong, the state engaged in gradual series of top-down reforms in information institutions that have slowly begun to bear fruit.19 As in Italy, the French government was under intense pressure to reduce its fiscal deficits to meet the Maastricht caps in order to qualify for the Euro, a high priority for the French government for several reasons. SIPO proceeds were a major factor in the success of the state in shrinking its deficit rate from above 5% in 1993-94 down to 2.9% in 1998. Ironically, the French state imposed most of the institutional changes that brought France into conformance with FPI expectations, even while decrying the growing presence of foreign investors and the threat they posed to French efforts at industrial structure policy and dirigisme in general. The changes in information institutions, and to a lesser degree in oversight and managerial compensation practices, are associated with France’s systematic program of SOE corporatisation.20 The German state was squeezed by the enormous costs of German unification in the first half of the 1990’s and by the Maastricht deficit cap in the second half of the decade. The fiscal deficit in Germany hit –3.3% in 1995 and –3.4% in 1996 - not as tight a squeeze as France, but still an incentive to liquidate SOE holdings of $15 billion over this period, equal to 4.4% of the budget. The biggest deal took place in three steps, between 1996 and 2000, when the state raised a total of $25 billion from the sale of 41% of the equity of Deutsche Telekom. The Deutsche 19 The China Securities Regulatory Commission (CSRC mandates that listed firms use a third-party auditor to verify financial statements. In 1993 the MOF issued a regulation on “Professional Qualifications of CPA Firms,” which began establishing professional standards under the aegis of the China Institute of Certified Public Accountants (CICPA) and formal licensing by the MOF. All Chinese firms listing on the Hong Kong and New York exchanges now use a Big Five auditor. China Securities and Regulatory Commission, Information Disclosure and Corporate Governance in China, paper for the OECD Conference on Corporate Governance, Hong Kong, pages 8-10; Martin Foley, “Accounting Adjustment,” The China Business Review, July-August 1998, page 23. 20 French listed firms, particularly SOE’s, were rapid adopters of IAS for domestic as well as international reporting: 40% of French public firms use IAS, including over one third of the CAC40 top firms. Third party audit is mandatory, and widespread, with the Big Five holding the dominant market share and a reputation for high professional competence. The state also pioneered the use of stock options for managers in corporatised SOE’s, a practice highlighted in “l’affaire Jaffré”, a scandal in which Philippe Jaffré, the head of Elf Acquitaine (an SOE), pocketed $56 million in options when Elf was acquired by Total Fina (another SOE) in 1999. Telekom transaction was listed on the New York Stock Exchange (NYSE) using Generally Accepted Accounting Principles (GAAP), triggering rapid adoption of the GAAP or IAS by leading German firms immediately thereafter and helping push two sets of financial reform legislation through the Bundestag, the KapAEG and the KonTraG.21 Moral Hazard Melt-Down’s The second event motivating these states to push governance reforms was the mounting price-tag of moral hazard losses in the financial sector during the 1990’s in most of these countries. Corporate governance failures in the financial sector were one of the root causes of these losses, which were ultimately picked up by the state. Prudential regulators wrestled with the peculiar corporate governance problem of financial institutions, especially banks, in all of these countries. Each flavor of insider governance poses its own risk to the state as lender-of-last-resort. Private blockholders are prone to use banks as a private piggy-bank to fund their other enterprises: this occurred in Belgium, Italy, Korea, Malaysia, Spain, and Taiwan. The state blockholder does such a poor job supervising managers (due to the owner’s incentive problem) that risk management fails and loan losses mount: this occurred in China, France, Korea, Malaysia, Singapore, and Taiwan. Banks controlled by employee-managers who are unaccountable to stockholders or an independent board also perform badly at risk management: this happened in Germany, Japan, and the Netherlands. Central bankers, prudential regulators, and finance ministerial officials in all of these countries realized during the 1990’s that outsider governance combined with prudential regulation (to counter related party lending by private blockholders) could substitute market discipline for failed insider governance. This motivated the state to undertake many of the reforms that boosted the compliance rankings in Table 4. For example, process-tracing of the burst of reforms in Korean corporate governance institutions in 1998 indicates that these reforms were rammed through by the newly-elected Kim Dae-jung government with a thin veneer of consultation with the blockholders, but essentially unilaterally by the state, 21 Germany’s company law (the Aktiengesetz) was modified by two comprehensive legal reforms, the Law on Facilitating Raising Capital (Kapitalaufnahmeerleichterungsgesetz, or KapAEG) and the Law on Control and Transparency of Corporations (Gesetz zur Kontrolle und Transparenz im Unternehmensbereich, or KonTraG), both passed by the Bundestag in early 1998. The KapAEG permits German firms to use IAS for their consolidated financial reporting rather than the rules set by the old Handelsgesetzbuch (HGB, or Commercial Code), thereby avoiding the cost of maintaining two sets of accounting records. under pressure from moral hazard losses during the Korean Financial Crisis.22 In a short period of time, the Korean government changed accounting to reflect the IAS, required third party audit (and increased the liability of accounting firms), enhanced disclosure requirements, and imposed the rule for a majority of independent directors on the boards of big firms, including the largest chaebol.23 Similarly, close examination of events in Japan suggests that moral hazard pressure was the prime mover behind Japan’s limited reforms in information institutions. The Ministry of Finance began to lose control over Japan’s monetary policy in the mid-90’s because of mounting concern over swelling bad debts in the banking system. As early as 1992, mid-level officials in the Ministry began to debate the need to tighten up the accounting system in response to mounting evidence of the moral hazard losses incurred during the excesses of Japan’s bubble asset inflation. In 1994, slack reserve accounting for potential loan losses allowed Japanese banks to report good profits and pay dividends even as their loan portfolios were deteriorating and both the real estate and stock markets continued to implode.24 A threshold was crossed in the last quarter of 1997, when a series of defaults by Sanyo Securities, Hokkaido Takushoku Bank, and Yamaichi Securities caused Tokyo’s interbank market to freeze up, as the magnitude of the hidden losses from securities trading and the unreliability of reported financial figures became clear to market participants.25 The Bank of Japan was forced to inject large 22 Non-bank financial intermediaries (or NBFI’s) controlled by the chaebol accumulated huge unhedged external liabilities during the 1990’s, liabilities incurred as they borrowed on foreign markets and re-lent to other members of the chaebol group, thereby demonstrating the risk of private blockholder insider governance of banks. These unhedged borrowings were one of the major causes of the 1997-98 Korean Financial Crisis. 23 In 1998 Korea’s newly launched Financial Supervisory Commission (FSC) overhauled Korean Financial Accounting Standards (KFAS), to bring them in line with the IAS, and external auditors were made mandatory for use by listed firms. New regulations also require detailed disclosure of financial transactions between chaebol blockholders and public firms, as well as independent, outsider boards of directors, in contrast to the former pro forma boards that rubber-stamped the orders of the founding blockholders. Chaebol were required to obtain 50% of their directors from outside, with a strict definition of “independence”. The legal obstacles to filing a derivative suit against a firm were reduced, from 1% to .01% of the stockholding, which made it much easier for small shareholders to bring such suits Regulations that limited hostile takeovers by capping such acquisitions at 10% of the target firm were removed, and voting procedures and tendering rules were changed to enhance the protection of minority shareholders. Il-Sup Kim, “Financial Crisis and its Impact on the Accounting System in Korea,” Korea Accounting Standards Board, KASB manuscript, January 2000; Bonchun Koo, "Corporate Restructuring and Financial Reform in Korea”, Korea Development Institute Working Paper No. 9807, 1998; People’s Solidarity for Participatory Democracy, “Shareholder Activism in Korea”, www.pec.pspd.org. 24 Takahashi Yoiichi, “The Truth behind Japanese Financial Crisis,” Princeton University manuscript, December 6, 1999. 25 This was made abundantly clear by third-party analysts, who rapidly downgraded the ratings of leading Japanese financial institutions and lead to a stinging “Japan premium” that these firms were required to pay in the international money markets. amounts of liquidity to keep the financial system functioning.26 MOF officials agreed that it was time to bite the accounting bullet; in any case economic recovery was stalled, and there was no end in sight for the banking crises. The MOF arranged for the Business Accounting Discussion Council (BADC) or Kigyô Kaikei Shingikai, to rapidly approve a series of GAAP principles for valuation and accounting, with special attention paid to accounting principles for financial institutions. Moral hazard pressure also accounts for the motivation of the German state to push for reforms of information and oversight institutions. A string of spectacular failures during the 1990’s (such as Metallgesellschaft) and the exposure of tampering with corporate balance sheets to gloss over financial problems (as in the case of Bremer Vulkan) demonstrated the need for increased transparency and monitoring of corporate accounts. 27 These bailouts had been expensive for the state; many of the write-offs were borne by Lander-owned banking institutions. As a result of these financial scandals and bank problems, all three political parties and Chancellor Helmut Kohl himself were increasingly suspicious of the oversight failings of the Big Three Banks. Reflecting this consensus, the KapAEG and KonTraG were passed with remarkably broad political support.28 This led to tightened financial reporting and accounting standards, a change in Germany’s oversight institutions, a limitation on the number of boards on which bankers could sit, and restrictions on banks’ right to vote proxies at their own discretion. Malaysia suffered moral hazard losses in the wake of the Asian Financial Crisis and the Malaysian state began its own program of top-down reform, pointing the finger at the governance sins of private blockholders. This top-down reform project began with a High Level Finance Committee on Governance established by the MOF in March ’98, which unleashed a series of regulatory changes through the Securities Commission, the Kuala Lumpur Stock Exchange (KLSE), and the Registrar of Companies.29 These changes led to the creation of a Malaysian Corporate Governance Code, the Malaysian Institute of Corporate Governance, and the Minority Shareholder Watchdog Group, whose power stemmed in part from equity holdings by government controlled funds.30 26 “Japanese financial markets clearly experienced a kind of credit crunch because of a rash of failures, declining asset prices, and growing mistrust of financial statements and regulators….That resulted in a further contraction of credit in what became a vicious cycle. In other words, unreliable financial statements had proved a serious impediment to the functioning of a market economy.” Mitsuhiro Fukao, Financial Crisis in Japan, October 26, 1999, Keio University manuscript, page 12. 27 “Arbeitsplätze durch attractive deutsche Kapitalmärkte,” December 12, 1999. www.cducsu.bundestag.de/texte/bericht. 28 “Intime Einblicke,” Der Spiegel, March 31, 1997. 29 “Crusade for Better Governance,” New Straits Times, 5/04/99, page 13. 30 “Minority Shareholders Watchdog Group looking for a Suitable Model”, Bernama News Agency, The Pension Funding Black Hole The third event motivating states to reform corporate governance practices was the yawning gap in the net present value of social security and pension schemes. Except for Malaysia and Singapore, the social security systems of the twelve countries in this sample are in the hole, with insufficient assets to cover future claims (payments to current beneficiaries are being covered by receipts from current workers). Table 10: The Public Pension Hole31 Country Net Present Value (% of 1994 GDP) Belgium -153 France -102 Germany -62 Japan -70 Italy -60 Netherlands -53 Spain -109 Table 10 shows estimates of the magnitude of this negative net present value of pension systems for seven of the countries in this sample. (China’s state pension system, like the banking system, is widely-considered to be insolvent, as are the pension systems in Korea and Taiwan.) Political leaders in these states all face the unpleasant choice of reducing benefits or hiking social security taxes, increasing the return on state pension assets, or replacing the bankrupt state pension system with a funded private pension system. Choosing either of the latter two options places the state on the side of FPI’s with regard to demanding minority investor protections. For example, the public pension systems in Malaysia and Singapore are fully funded, the legacy of an enforced provident savings scheme. As noted earlier, state-controlled pension plans served as a potent tool for implementing the top- down corporate governance reforms in those two countries and help account for the high conformance levels.32 Singapore’s large net external asset pool in effect 8/16/99. 31 Roseveare, Liebfritz, Fore, and Wurzel, Aging Populations, Pension Systems and Government Budgets: Simulations for 20 OECD Countries, OECD Working Paper No. 168, 1996, Table 1. 32 Singapore’s Finance Ministry launched a three-pronged effort to modify Singapore’s governance institutions by setting up a Committee on Company Legislation to reform the company law and control issues, a Committee on Disclosure and Accounting Standards to review information institutions, and a Committee on Corporate Governance to focus on oversight, plus a new Singapore Institute of Directors - transformed the state into a large FPI, with similar preferences for improved minority investor protections around the world. State-controlled pension funds are the largest single investor in the Singapore stock exchange, with Temasek holding an estimated 25% of the total equity market capitalization, and the state holds an estimated $100 billion in offshore assets, including a large block of equity. The decision by the Netherlands government to privatize and fully-fund its pension plans provided one of the few counterweights to resistance against improved governance from Netherlands managers in league with labor unions. With a record of conservative fiscal management and an average fiscal deficit among the lowest in the European Union (EU), the Netherlands state did not come under the sort of budget pressure for privatization of the state sector that occurred elsewhere in Europe. Nor did the Netherlands experience the moral hazard losses that troubled other governments. As it entered the 1990’s, however, the state faced a severe pension crunch due to unfavorable demographics and the richness of its retirement benefits. Aware that its social security taxes were among the highest in Europe, and facing resistance from Labor on trimming of benefits, the Netherlands state chose to privatize its state pension scheme in 1990, seeking higher returns from the large pool of pension assets that had hitherto earned low returns under state management. It also removed the caps on equity exposure and international exposure from these funds. The improvement in the return on these assets is one of the reasons why The Netherlands now has a relatively low pension deficit ratio among EC countries, 53% of GDP. As a result of these pension reforms Netherlands pension funds rapidly transformed into return-sensitive investors similar to those in the U.S. and the U.K. They banded together to confront poorly-performing Netherlands companies at Annual Meetings, in an informal Association of Pension Funds, and collectively hired Déminor, a third-party governance consultant, to subject large Netherlands companies to close scrutiny. They challenged the structuur- regime on several fronts, although on this point the former state pension funds were more aggressive than the corporate pension funds, which were reluctant to criticize their parent firms. By the same token, attempts by the French state to improve its pension system strengthened the hand of the Association Française de la Gestion Financière (AFG-ASFFI), the association of money management and pension firms, which became a vocal critic of entrenched managers and inefficient SOE’s. AFG-ASFFI staffed by officers from GLC firms. “Move to improve corporate governance”, Business Times Singapore, 12/99, page 2. urged the government to reform France’s pay-as-you-go pension scheme with a fully-funded quasi-private system, even as it took the lead in pressing French firms for improved corporate governance, including improved disclosure, more NED’s, and removal of unequal voting rights.33 AFG-ASFFI issued its own governance code and funded third-party scrutiny of French firms on a collaborative basis through France’s two leading third-party fee-based analysis firms, Déminor and Proxinvest.34 The German state has been slowly coming to grips with its pension funding hole, but, as in France, with resistance by organized labor unions and by employee- managers. Again, the most vocal proponents of a fully-funded private system - and enthusiastic supporters of governance reforms - were money management and pension firms, financial analysts, and the shareholder’s association. For example, the German guild of financial analysts, the Deutsche Vereinigung für Finanzanalyse und Asset Management, which represents one thousand professionals in four hundred investment banks and asset management firms in Germany, issued a set of formal guidelines that “scorecard” the accounting, auditing, and disclosure practices of listed firms, thereby reinforcing the assignment of a corporate governance premium and generating more pressure for reform.35 Japan’s mixed public/private pension system is also in a deep hole; private Japanese firms have a huge unfunded pension liability to their employees, estimated at somewhere between 60 and 80 trillion yen.36 The portfolio returns of the mixed private-state pension system, as well as the plans managed by Japanese insurance and trust banks (shintaku ginko) have been dismal, with a current yield of about 2.5%, 3% below the return needed to meet the actuarial commitment to pension beneficiaries. The Ministry of Health and Welfare, which supervises these pension plans, issued a series of administrative regulations that tighten the fiduciary obligations of pension trustees and require them to maximize asset return (rather than other stakeholder values) and to focus on, inter alia, good corporate governance of the firms in which they invest.37 33 Alain Leclair, “The French Experience: An Activist Corporate Governance Strategy”, paper for the International Corporate Governance Network, Tokyo, July 2001. 34 “Deminor rates 300 European Companies based on Corporate Governance Standards”, www.deminor.com. 35 Deutsche Vereinigung für Finanzanalyse und Asset Management, “Scorecard for German Corporate Governance,” July 2000, www.dvfa.de. 36 Goldman Sachs Japan, Portfolio Strategy: Tsunami Alert, October 16, 1998. 37 Kôseisho, Nenkin shikin unyô bunkakai, “kôsei nenkin hoken oyobi kokuminnenkin no tsumetatekin no unyô ni kansuru kihonhôshin ni tsuite”, February 2001; Tomomi Yano, “Nenkinshikin unyô kara mita corporate governance”, paper for International Corporate Governance Forum, Tokyo, July 2001. Since revised accounting principles require firms to disclose their unfunded liability, an exodus began (estimated at thirty trillion yen out of a total of fifty trillion yen) away from the mixed public/private pension system into privately managed funds.38 As a result, the market share of foreign investment advisory firms (who behave exactly like FPI’s) has grown from zero to 10% of Japan’s pension market in the last ten years. Preserving the Public Benefits of Control Overall, the pattern of change in institutions in this sample reflects the governance preferences of politicians to preserve the public benefits of control while striving to manage the fiscal consequences of privatization, moral hazard losses, and pension underfunding. The state could reap SOE SIPO’s on the basis of improved accounting and audit practices, with nominal changes in oversight, but without relinquishing control of these firms to value-maximizing outside directors. The employee-managers and directors of these firms (bureaucrats or political appointees) were compensated with stock options – pioneering the use of stock options in many of these countries. Hostile takeovers by a value- maximizing acquirer were blocked by anti-takeover and voting rule obstacles. In all twelve countries (except China) SOE SIPO’s were accompanied by careful consultation with labor unions. This is consistent with the empirical literature on privatization, which indicates that post-privatization lay-off’s are limited to either “transition” or lower-income developing countries.39 Nor is there much evidence that agency or expropriation costs declined after SIPO’s.40 Though the literature suggests that there are substantial performance improvements in post- SIPO SOE’s, including greater output per employee and reduce leverage, it is not clear that this is due to lower expropriation costs by the state blockholder. In every country example, the state adopted changes in information institutions, in part to obtain the revenues from SOE SIPO’s, but drew the line at changes in oversight practices or control institutions that would open these “privatized” firms to the discipline of a market for control. There is little evidence that post-SIPO SOE’s have boards composed of independent outsiders; perusal of a random sample of SOE share offering memoranda reveals boards filled with serving or retired government officials, politicians, and labor union representatives, with a small sprinkling of managers of large firms. More important, in this sample the state retained, on average, at least 50% of the voting rights in SOE’s after the SIPO’s and several rounds of 38 At an exchange rate of ¥110/$1, 30 trillion yen equals approximately $270 billion. 39 William Megginson, “From State to Market: A Survey of Empirical Studies on Privatisation”, Journal of Economic Literature (forthcoming), pages 32-34. 40 Megginson, op. cit. secondary offerings. This commanding blockholding, combined with various “golden share” and formal veto rights, effectively buffers these firms from any chance of a successful contest for control. Process-Tracing Proof for the Public Purpose Model In sum, process tracing of actual regulatory changes points to the state as the prime mover of governance changes, and lends credence to the public purpose model of governance change. Once these state-imposed reforms are in place, such as better accounting, auditing, and disclosure, they serve to reduce the private benefits of control extracted by private blockholders, by making it more difficult to extract those benefits – increasing the cost of stealing, and imposing, as it were, a “tax on blockholders”. Entrepreneurs or private blockholders who are strapped for cash and thus must access equity markets for capital, tend to became “early adopters” of good governance, thereby increasing the competitive pressure, over time, on incumbent blockholders and entrenched employee- managers to adopt similar reforms. There is some evidence that the parallel stock exchanges for entrepreneurial companies that were established in several countries in this sample, including the Neuermarkt in Germany and the KOSDAQ in Seoul, had slightly better corporate governance practices than the major stock markets (although this impression did not often survive the collapse of the high-tech bubble). Thus the endpoint of the public purpose model is the same as the private markets model, but by different pathways, with different actors in the vanguard. The FPI good governance premium offered a direct price incentive to the state with regard to the governance practices of SOE’s, and also set in motion an indirect premium for better governance by means of domestic pension funds. The causal connection between FPI and moral hazard losses is more attenuated, and the costs of poor governance were largely endogenous to each country, although international capital markets persisted in highlighting these costs, much to the discomfort of finance bureaucrats and politicians alike. Implications of the Public Purpose Model The public purpose model of governance convergence, if accurate, has several implications for theory of political economy as well as public policy. Governance change in any given country depends primarily on how badly the state is squeezed by the three fiscal pressures, how quickly private blockholder “early adopters” forgo private benefits, and how much resistance to governance reforms the state must overcome from private blockholders and employee- managers. This predicts country-specific path dependence in the sequence and rate of change. Instead of uniform convergence, it suggests the pattern of governance change will persistently vary between countries, as the fiscal pressures can present themselves in different combinations and intensities over time. The public purpose model confirms the suspicion of some governance economists that private benefits of control are stubbornly resistant to change. There are steep obstacles to the process by which private blockholders trade their private benefits of control in exchange for higher stock valuation. By the same token, politicians and bureaucratic elites who control the state may find it as difficult to relinquish the private benefits of control of state-owned firms as do private blockholders. The pivotal role played by the benefits of control suggests that the scholarly debate over whether there are one, or many, versions of capitalism may be clarified by viewing institutional formation through the more parsimonious lens of ownership type. Ownership determines where the potential income cleavages run in response to international inducements and pressures, and the shifting terms of state ownership are a key determinant. In order to predict how these cleavages affect policy outcomes, political science and international relations theorists may have better results using the ownership hypothesis when examining the effect of exogenous forces on domestic choices.41 A great deal of ink has been spilled in the political economy literature creating elaborate taxonomies of “stakeholder versus shareholder capitalism”, “producer versus consumer capitalism”, and even “Nippo-Rhenish versus Anglo-American capitalism” in order to explain contrasting institutions and policy outcomes. However, these findings suggest a mundane question of “who gets the money?” may provide a more parsimonious explanation.42 In an early study of how external economics affects domestic institutions, Zysman focused on the structure of the banking system as the key explanatory variable, arguing that “by knowing the financial system one can predict the nature of the process of adjustment.”43 These twelve cases suggest the paraphrase “by knowing the ownership type one can predict the nature of the process of adjustment.”44 41 See Robert Putnam, “Diplomacy and Domestic Politics,” International Organization 42 (1988), 427-60 and Helen Milner, Interests, Institutions, and Information: Domestic Politics and International Relations, 1997. 42 See Berger and Dore, National Diversity and Global Capitalism, 1996; Hollingsworth and Boyer, Contemporary Capitalism: the Embedded ness of Institutions, 1997; and Michel Albert, Capitalism vs. Capitalism, 1993, all of which speculate on institutional divergence or convergence as a function of many variables, including the type of state, but not as a function of ownership. 43 John Zysman, Governments, Markets and Growth: Financial Systems and the Politics of Industrial Change, 1983, page 91. 44 Traditional political science terms such as Business and Labor are poor predictors of who wins and who loses in response to exogenous influences such as FPI. As demonstrated by these twelve cases, “Business” As for the debate on whether non-state actors matter in policy outcomes, the corporate governance example proves unambiguously that they do: the state’s mediation of financial inducements offered by FPI determine the shifting pattern of corporate governance around the world. As a result of the FPI tidal wave and higher ratios of stock market valuation to GDP in all of these countries, the state and foreign investors are locked in an uneasy embrace that either party would find expensive to relinquish.45 The corporate governance example suggests that convergence to a global standard of governance is not a “race to the bottom,” in which the state that deregulates fastest wins by attracting the most capital at the lowest cost. 46 On the contrary, the state with the most impartial, efficient protection of minority shareholder rights is the one that wins by attracting the most capital at the lowest cost. The spread of Anglo-American style corporate governance is not equivalent to deregulation, for capital markets are embedded in a set of sophisticated formal and informal institutions: in many respects the market-ordering Anglo-American model requires more institutional support than private-ordering or state- ordering models.47 Nor is it a simple process of re-regulation, whereby the state uses a different set of tools to pursue the same objectives, as interventionist bureaucrats attempt to harness equity markets to their existing developmental model.48 Instead, it results in para-regulation, with a shift of authority to investors can represent any one of three different ownership groups – private blockholders, state blockholders, or employee-managers – with radically different institutional preferences. 45 The state cannot sit down with FPI’s and “hammer out a deal” regarding the terms and conditions of corporate governance. This contrasts to bank debt markets, in which the MOF can sit down with a dozen or so big banks and work out an agreement on, say, debt swaps in a crisis. It is also unlike FDI markets, in which the state can summon the investor and negotiate (from a position of considerable strength) the terms of the deal. FPI’s face the inverse problem of collective action in responding to the state. There are many of them, with heterogeneous preferences. They are limited in their ability to flee from a given country: rapid exit incurs high transactions costs, especially during a financial crisis, and they are increasingly constrained in exit by the practice of portfolio indexing: for example, the top 25 U.S. institutional investors use indexing to manage 60% of their equity portfolio. Once committed to such an index, for example, investors are predisposed to leave a given percentage of their portfolio in a given country market: if they don’t like the corporate governance standards, they must work to improve them. 46 Dani Rodrik, Has Globalization Gone Too Far? 1997. 47 The complexity of the Anglo-American governance model should give pause to the conventional wisdom assumption that the Anglo-American model is easier to imitate than Japanese or German models, which require “idiosyncratic” governance institutions. See Bernard Black, “Creating Strong Stock Markets by Protecting Outside Shareholders,” paper for the OECD Conference on Corporate Governance in Asia, March 3-5, 1999, pages 6-7. 48 Steven Vogel, Freer Markets, More Rules: Regulatory Reform in Advanced Industrial Countries, 1996. above the level of the state, combined with a greater shift of authority to independent statutory regulatory agencies below the state level. 49 Explanations of institutional changes in response to global markets do not require the invocation of “American hegemony” that frequently pops up in discussions of financial globalization.50 The spread of the Anglo-American model of corporate governance has little or nothing to do with policy goals of the United States government; FPI markets and sovereign states abroad are the key actors in bringing about governance changes, at least in this twelve country sample. Any influence acquired by the state in Washington or London as a result has been obtained the same way that Britain acquired its global Empire, “in a fit of absentmindedness” rather than by design.51 From a policy standpoint, “reforms” to corporate governance institutions, particularly in emerging markets, must be careful to deal with agency and expropriation costs in balance. Governance institutions ex ante the introduction of FPI are complementary, and they rely on each other to mitigate, if not eliminate, both expropriation and agency costs.52 Changing just one institution at a time, without examining the question of complementarity, may have undesirable effects. For example, efforts to reduce expropriation costs by improvements in NED oversight, which in effect levy a heavy “tax on entrepreneurs”, may blunt the ability of private blockholders to discipline managers. The unintended consequences of these reforms can be soaring agency costs – with entrenched employee-managers now free to manage badly. Indeed, in the corporate governance debate in some countries, calls to “professionalize” managers are often a code phrase for “reduce the influence of blockholders.” Conversely, efforts to reduce agency costs - changes that open the door to free-for-all takeover contests - may result in high expropriation costs, as blockholders emerge to discipline (or expropriate) employee-managers and, in the process, 49 The Anglo-American model is based on securities regulation by quasi-independent agencies such as the SEC, accounting rules by semi-private bodies such as the FASB, and enforcement of minority rights and fiduciary responsibility by independent common-law courts, rather than by Ministries of Finance or Ministries of Justice. 50 See for example Robert Gilpin, The Political Economy of International Relations, 1987, and The Challenge of Global Capitalism, 2000, or Loriaux, op.cit. 51 "We [the English] seem, as it were, to have conquered and peopled half the world in a fit of absence of mind." Sir John Seeley, The Expansion of England, 1883. 52 Heinrich,Ralph. 1999. "Complementarities in Corporate Governance: A Survey of the Literature with Special Emphasis on Japan," Kiel Working Paper No. 947, Kiel Institute of World Economics, Kiel, Germany. compensate themselves for these efforts at the expense of minority shareholders.53 In terms of industrial organization, convergence may halt at shallow equilibrium, with potentially unpleasant consequences for product/market competition. States are extremely sensitive to the potential fall-out from hostile takeovers. Even if states gradually adopt a para-regulatory approach to supervising corporate governance, this will not necessarily change the political disincentives associated with market-based contests for control, or make these states more willing to expose corporatised SOE’s to such contests. As noted earlier, the pattern of conformance has stopped well short of accepting market-based contests for control in ten out of twelve of these cases. 54 States’ distaste for such contests, particularly for the corporatised SOE’s that now account for such a large weight of their stock market capitalization, suggests that one result of the SOE corporatisation trend in the late 1980’s and 1990’s may be higher agency costs. Because of remaining state majority control, or because of voting restrictions or “golden shares” imposed by Ministries of Finance, the managers of these firms are likely to become more entrenched and less disciplined by the threat of hostile takeover’s, while strengthening their “stakeholder” common cause with the rest of the SOE workforce – especially if that workforce is unionized and politically mobilized.55 They may also engage in classic empire-building by acquiring other firms at above-market prices, while being protected from such take-overs themselves by the state, and insulated from the financial market discipline that otherwise makes such acquisitions dangerous to the employee-managers.56 53 Blockholders can expropriate employees by breaking implicit contracts on the sharing of the firm’s “quasi-rents” or simply by looting their pension funds (as in the Robert Maxwell scandal). See Margaret Blair, Ownership and Control: Rethinking Corporate Governance for the Twenty-First Century, 1995. 54 The two frequently-invoked examples of hostile takeover in Europe, the Vodafone-Mannessmann transaction in Germany and the Olivetti-STET transaction in Italy, may be an aberration and not indicate a fundamental change in those two countries. 55 The “grabbing hand” model developed by Shleifer and Vishny suggests that in many cases politicians can extract as many private benefits from SOE’s that are partially or even fully privatized as long as they can retain control over the profits of the firm through discretionary regulations. Depending on the degree to which corruption is tolerated, they will take these private benefits in cash or in terms of excess employment for favored unions or regions. Shleifer and Vishny, The Grabbing Hand, pages 176-78. 56 “Golden shares prevent that bad bidders become good targets. In an unconstrained control market an acquirer that failed to achieve synergy would run into difficulties and would become a prey of other companies. Golden shares protect potential bad bidders. Partial public ownership also protects potentially bad acquirers, since governments may value control of the company for political reasons more than cash flows derived from selling ownership rights. There seems to be an asymmetry in that some firms still controlled by the public sector (through partial ownership or golden shares) are active acquirers abroad while their governments keep important restrictions at home. The Spanish government vetoed the merger between KPN and Telefónica and the Italian government vetoed the merger between Telecom Italia and Deutsche Telekom.” Francesc Trillas, Mergers, Acquisitions, and Control of Telecommunications Firms in Europe, paper for the Regulation Initiative, August 2000, www.london.edu/ri, page 7. The minority shareholders of these firms, domestic and foreign alike, are unlikely to be cheerleaders for continued deregulation, if it poses a threat to the earnings (rents) of the largest stocks by value. This may lead to a new “shallow equilibrium”, with states and FPI’s aligned with entrenched employee-managers - all three of whom now resist further deregulation and competition, and jointly collect respective portions of the rents.57 In this way, an unintended consequence of SIPO’s and increasing FPI may be a roadblock erected in the path of continued product-market deregulation, while encouraging inefficient acquisitions and cross-border mergers. This process could quickly become politically contentious, not just domestically, but across borders as well. There is already considerable political fallout within Europe as corporatised SOE’s spar with each other in attempts at cross-border acquisition and consolidation. The process of consolidation among “national champions” is even more delicate in Asia, as witnessed by the squabbles between Singapore Telecom, Malaysia’s Time Telecom, and Hong Kong Telecom. In both regions, hostile cross-border contests can rapidly escalate into high politics. At that point, states may realize that they have adopted, more or less by default (if not in a “fit of absentmindedness”) a set of market-driven Anglo-American corporate governance institutions without having forged the domestic political consensus to deal with the consequences. However, demographics and increased competition in financial markets will keep the process moving forward, albeit in fits and starts. As states begin to fund their pension systems, and as households and firms assert more control over their pension assets, this places more pressure on pension fund managers to seek higher returns domestically and abroad. These changes in pension plans will mobilize a huge amount of previously passive savings into active equity investment, especially in Japan and continental Europe. Active equity investment leads to a more systematic use of the governance discount (or premium). This suggests that regulations regarding the structure and function of pension plans are a central issue of public policy, with enormous long term consequences for equity markets generally and for corporate governance specifically. Pension plan reforms may turn out to be the tail that wags the corporate governance dog. 57 A shift on either the supply or demand side of this equation could upset equilibrium at shallow convergence. On the supply side, FPI’s could tire of swelling agency costs at corporatised SOE’s, driving the stock price of these firms to embarrassing lows. On the demand side, fiscal pressure on the state could force it to liquidate the balance of its equity holdings in these firms. Either factor could lead to a contest for control of these SOE’s, whereby private blockholders buy out the state’s interest in private transactions, gobbling up bits and pieces of the former “national champions” and natural monopolies, or engaging in nasty public contests for control. Retrospectively, the explosive growth of assets managed by institutional investors in the United States and the United Kingdom during the three decades between 1970 and 2000 caused Anglo-American investors to became serious players in international equity markets during the 1990’s, thereby providing the price incentives for adopting the Anglo-American model of corporate governance.58 Looking forward, as states increasingly permit domestic savings to freely enter the FPI pool, these minority shareholder protections will lose their “Anglo-American” appellation, the para-regulatory practices that support these protections will spread, and countries may be impelled from shallow to full convergence. James Shinn Jshinn@cfr.org October 15, 2001 Discussion Paper 58 This growth was driven by tax changes (such as 401-k plans) and rules regarding private pension plans such as the ERISA statutes.
Pages to are hidden for
"Equity Capitalization"Please download to view full document