1
3/2/16 world bank , Gourevitch revisions
Draft notes: not for quotation or attribution without permission of author.
“Politics, Institutions and Society:
Seeking better results.”
Peter Gourevitch
UCSD, La Jolla Ca.
For 2005-06 Russell Sage Foundation
Prepared for
World Bank Legal Forum 2005 on “Law,
Equity and Development”
For session on
“Building Equitable and Inclusive Markets.”
The World Bank
Washington, DC
December 1-2, 2005
Revised May 2006
2
3/2/16 world bank , Gourevitch revisions
Abstract
We know political processes influence growth and equality, but we disagree on just how.
The differing camps in combat can be sorted into three categories of causality: legal
family ; political institutions, or de jure institutions; and social forms of power, or de
facto institutions. The first stresses the impact of contracting arrangements generated by
civil and common law systems, whereby the latter are held to be more encouraging of
growth; the second stresses the impact of political arrangements, whereby structures
obligating accountability of governments to constituencies inhibit predation and
encourage growth; the third stresses the ability of social forms of power (land, media,
family, organizational capacity) to bend governments to their will, whereby high
concentration of social power undermines the capacity of institutions to restrain
predation. These arguments are examined with the specific example of debates about
variance in systems of corporate governance: why some countries have patterns of diffuse
shareholding while others have concentration and control by blockholders. The legal
family school has difficulty explaining variance within countries over time, and variance
among countries within the same legal family; the formal political institutions school
underplays the role of social forms of power among interest group; the social power
school underplays the role of institutions in explaining variance among unequal societies.
We have better measurements of the first two categories (formal law and political
institutions) than we do of the third. A key ingredient for promoting economic
development is consider policies that strengthen those social groups that seek it.
3
3/2/16 world bank , Gourevitch revisions
We know that political processes influence growth and equality. There is indeed a
substantial consensus on this, a marked change from the recent past. Indeed so great is
the consensus on the importance of politics, that we are able now to disagree on the next
big point: just what model of politics do we have in mind?
The latest World Development Report 2006 Equity and Development is
commendable for exploring political and social variables in shaping policy
determination and actual outcomes. Here I wish to draw out some of the issues that
appear in discussions of politics. I will use examples from the field of corporate
governance: Why do corporate governance systems vary around the world ( diffuse
shareholder models in the US and the UK,
concentrated blockholder models in most
of the world), and what can be done to improve these systems?
Once a topic unto
itself, the debate about corporate governance outcomes now tightly interweaves with
important discussions about
economic development and “endogenous institutional
change.” What produces “good political institutions” that cause economic growth (and
along the way, good corporate governance, equity, freedom and justice)?
In these debates about explanation, three arguments contend for primacy: a) the
formal rules of law (legal family), associated with the research of La Porta, Lopez di
Silanes, Shleifer , Vishny1) ; b) political institutions, or de jure, parchment institutions,
associated with constitutions2 ; and c) social forms of power, elements of civil society,
such as interest groups, social class, media , professional associations, or de facto
institutions.
4
3/2/16 world bank , Gourevitch revisions
Each of these arguments has a different sense of “politics” and political
mechanisms. For the legal formalism school, good institutions that promote growth and
corporate governance come from legal family, where common law is superior to civil
law; improving country performance means improving the codes that regulate corporate
and securities law. The political institutions school argues that the implementation of
codes, as well as passage of “good” laws, requires support from policy making
institutions; “good “ institutions produce good policy, so change has to come from
improving political decision-making bodies. The political sociology school argues that
institutions themselves express the preferences of groups in society who actually use the
institutions and can bend them to their will; good policy requires altering the distribution
of power in society. Thus the debate is among three forms of power --legal, institutional
, and social -- and thus three models of politics.
The rest of the paper explores the debate through the specific prism of corporate
governance. The next section sets out corporate governance as a dependent variable: the
outcome to be explained, the variance we observe around the world in patterns of
corporate governance. Then, the paper looks at the three types of argumentation. Finally,
we explore implications for policy and the larger question of development promoting
institutions.
I.
Variance in Corporate governance outcomes.
Corporate governance systems are contrasted by the degree of shareholder
concentration. In the blockholder models, a few actors (banks, governments,
individuals, or other firms) hold a large percentage of the shares. This allows them, as
owner principals, to monitor directly, or through agents on the board, the behavior of
5
3/2/16 world bank , Gourevitch revisions
their agents, the managers. They have the means to act ( controlling majority of seats)
and the motive (a large ownership share). By contrast, the diffuse shareholder model
elects a board of directors who represent the interests of these shareholders without any
direct action by them. Minority shareholders are protected in the diffuse system from
insider dealing by blockholders, but lack the means and motive to intervene directly or to
monitor their mediating agents ( the elected directors) . Minority shareholders are
vulnerable to opportunism by insiders in the blockholding model, but can also free ride
off the activism of those blocks in watching the managers.
Comparative analysis thus begins by describing the difference in shareholder
concentration diffusion.
3
Drawing on the foundational work of the La Porta et. al, the
World Bank, and many others, , Gourevitch and Shinn show national differences in in
Figure 1 , “Ownership Concentration” by country 4;
6
3/2/16 world bank , Gourevitch revisions
7
3/2/16 world bank , Gourevitch revisions
We see considerable variance across countries While Japan and China are at the
top, these merit important qualifications. In China only a small portion of the economy
lies in listed stocks, the rest is state owned. In Japan, we find extensive cross
shareholding, so that there can be no effective market for control. No one would regard
either country as exemplifying the diffuse shareholder model. Thus the most diffuse
country is the US, the least is Chile.
There is not substantial disagreement on this description. The battle heats up over
explanation. Why does this divergence in blockholding take place?
II.
Explaining the Variance: Three Models of Politics
We can sort the contenders into three camps: legal family ; political institutions;
and interest groups. They differ most on what shapes law and regulation: If law and
regulation protect investors from exploitation by insiders, what explains the supply of
8
3/2/16 world bank , Gourevitch revisions
these laws and regulations?. There is an additional debate over whether law matters and
which laws matter; we incorporate that debate into our discussion below.
A. Legal family – the formal rules approach.
For diffusion to occur, an investor must accept a minority position in the firm , as
outsiders coming in , or as blockholders selling down. In so doing, these investors risk
exploitation by insiders. To so, argues a rich literature , they must obtain some kind of
protection from insider abuse. Law and regulation provide this -- Minority Shareholder
Protections.5 These measures include: information (disclosure and audit), oversight
(board independence), control rules (voting processes), and managerial incentives
(executive pay). The higher the level of MSP, the more supervision of managers, the
more reassured are investors (who buy) and blockholders (who sell to diversify) and the
deeper will markets become --at least that is the core assumption of the “law matters”
school.6 Laws about insider trading, the composition of boards, the issuance of shares,
providing information about the firms finances, take overs and the market for control –
these allow outsiders to judge the value of their investment and take action: buy, sell, or
take over.
The prediction of this approach is that where MSP is strong, diffusion occurs. .
Comparative research measuring various components of MSP was pioneered by LaPorta ,
et al, thereby helping to create a field of study. Initially they measured issues about firm
governance , but they and others7
have broadened the terrain to include other elements.
Our index, Table 1, combines several measures , drawn from these authors and our own
evaluations.
9
3/2/16 world bank , Gourevitch revisions
(NOTE TO EDITORS: Please correct the label of this table to Table 1:
Table 1 Minority shareholder protections index
10
MSP index G/S
3/2/16 world bank , Gourevitch revisions
11
3/2/16 world bank , Gourevitch revisions
In the legal approach, high levels of MSP should correlate positively with
shareholder diffusion: the stronger the MSP, the more investors are willing to be
minority shareholders. Figure 2 correlates Blockholding with MSP.8
Figure 2 Blockholding and Shareholder Protections
12
3/2/16 world bank , Gourevitch revisions
So, MSP and corporate governance patterns do connect, as expected. Some
relationship exists – though the correlation is not strong, which suggest other variables
are at play. We come back to these other variables, but first we address the “hot spot” of
arguments about politics. To the extent MSP (law and regulation about shareholder)
accounts for diffusion, what explains the provision of MSP? La Porta et. al argue it is
“legal family” – whether a country has civil or common law. Common Law systems,
they suggest, have higher MSP than do Civil Law countries. Having measured MSP in a
large number of countries, La Porta et. al then correlate levels of MSP with legal family.
La Porta et. al provide evidence that confirms the intriguing “eyeball” observation that
the UK, and its former colonies including the US, are heavily grouped in the shareholder
diffusion and high MSP quadrant, while France, Spain and their former colonies tend
toward the blockholder and low protection pattern. This is a striking pattern and deserves
thought.
And yet, closer inspection reveals problems with the legal family explanation.
On the empirical side, we observe substantial variance in corporate governance patterns
within countries over time, while their legal family stays constant, and we observe
substantial variance among countries within the same legal family. In their important
paper, “The Great Reversals ” 9 Rajan and Zingales track within -country change
extensively, showing France and Japan developing considerable stock markets prior to
WW I, then showing a sharp swing toward bank dominance. The US looked much like
Germany prior to WWI, but policy changes created the regulatory system of anti-trust,
banking regulation and MSP which produced the system we know today. 10 Berkowitz
and Pistor show a “transplant effect” , that local conditions influence considerably how
13
3/2/16 world bank , Gourevitch revisions
countries alter efforts to import legal structures from outside.11 Within the same legal
family, considerable variance exists among countries in what they do.
On the conceptual side of the legal family argument, the mechanism of causality
remains unclear: just why does legal family produce the claimed effect?
Legal family, a formal characteristic, does not explain the content of the actual laws
which are passed, and , even more important, the pattern of enforcement. Nothing about
common law guarantees that countries will enact vigorous MSP’s , nor that they will
enforce them. Nothing about civil law systems prevent countries from enacting strong
MSP, nor from enforcing them.
Laws are passed and sustained because of political processes: the demands of
various groups in society and the institutions that process them. It is to these we turn.
B. Political institutions – de jure power: Laws are made, and enforced, by
political systems. Variation in the content of laws, regulation and enforcement, might
then be the product of variation in political systems. In this respect, formal legal system
( civil vs. common law) can be differentiated from political institutions and from social
structures.
Acemoglu , Johnson and Robinson call the legal systems “ contractual
institutions,” while labeling political structures as ” property rights institutions.”12
Contractual processes, they argue, do differ according to legal family, but do not predict
economic development. Property protection institutions don’t correlate with legal family
and do correlate with economic growth. We find similar results for corporate
governance.
What are property protecting institutions? In formal terms, they are the structures
which organize power in political systems. Here Acemoglu , Johnson and Robinson
14
3/2/16 world bank , Gourevitch revisions
make another quite important distinction: between de jure institutions, formal structures
that shape the way “preferences” (interests, ideas, goals) are aggregated, 13 and de facto
institutions (forms of power in civil society , such as land, guns, money. We know that
different methods of counting the votes yield different outcomes because of cycling and
agenda control. The same applies to other aspects of political system. This insight
generates an important literature comparing countries along different dimensions of
their political institutions, including electoral laws, the structure of the executive and
the legislature,
and press.
federalism, courts and the independence of the judiciary, free speech
Countries are compared with categories such as Presidential vs
Parliamentary systems, Majoritarian vs. Consensus systems, high or low number of veto
gates or veto players.
Important World Bank research by Beck, Kiefer and others has
made an important contribution to this literature, for a number of years, building a
valuable data base
14
; other research projects have added substantially to our
knowledge.15
The institutionalist approach – the focus on de jure institutions-- provides
leverage in accounting for variance in MSP provision and corporate governance
outcomes. In stable democracies, for example, Gourevitch and Hawes and Gourevitch
and Shinn 16 show that MSP and corporate governance outcomes correlate with the
Majoritarian/Consensus distinction. Majoritarian systems are more likely to have high
MSP and diffuse shareholder patterns. The logic has to do with the impact institutions
have on the bargains among groups in each system. Majoritarian systems have wider
policy swings, which undermines the corporatist foundations of a particular set of
arrangements around blockholding, which then induces actors to seek autonomy through
15
3/2/16 world bank , Gourevitch revisions
arms length economic relationships, and thus to MSP and shareolder diffusion;
conversely, “Consensus” systems reinforce the bargains in corporatism and sustain
blockholding arrangements.17
Pagano and Volpin18 stress the impact of electoral
systems: countries with proportional representation are more likely to have high
shareholder protection rules and diffuse shareholding patterns than are countries with
once-past the post single member plurality districts.
The impact of veto players and stable institutions may be different between
stable democracies and authoritarian regimes. Strong regimes can enforce rules but can
also predate on economic life, undermining incentives to invest; weak regimes can’t
enforce rules, which also undermines incentives. Blockholding may arise out of either
extreme. Having many veto points can create policy paralysis. Having few veto points
lessons the chances of policy paralysis, but increases the threat of predation. The
confidence of investors may be highest at some intermediate point on a U shape curve,
some “just right” balance, between inaction and too much. We do know that for some
periods, investment in authoritarian regimes can be high; such regimes are able to signal
some kind of commitment to investors that predation will be limited.19
C) Social forms of power, de facto power. In contrast to this stress on de jure
political institutions, another literature looks at de facto forms of power – control of land,
guns, people—which enables social groups to manipulate the political process to their
advantage. 20 Constitutions seek to constrain later generations by structuring the political
game on behalf of whoever is writing the documents; the constitution is thus a
commitment techology. But groups with social power are able to undermine these
restrictions, capturing the formal structures, altering their usage, or even changing the
16
3/2/16 world bank , Gourevitch revisions
constitution.
The history of Latin America, Africa, indeed Europe and the US are full
of examples of this kind. Acemoglou and Robinson cite the American South after the
civil war, Colombia, and Liberia as examples where landed elites blocked formal
political efforts that sought to limit their power. In the American South, the 14 th
amendment providing equality to freed slaves was rendered meaningless by cutting the
voting rolls, terrorism from the Klu Klux Klan, and Jim Crow exclusion from education,
jobs and many social practices. In Colombia, the landed elite made a bargain to rotate
power and exclude most of the population from participation in politics. In Liberia, the
descendents of American slaves did the same. In France, the authors of the Third
Republic’s Constitution planned for a strong executive hoping for a return to monarchy.
When the President lost a fight with the National Assembly, the system developed rapidly
into Parliamentary dominant, weak Presidential model.21 Stasavage critiques the
influential North Weingast essay on the English political settlement of 1688, by noting
the limits on the crown came not only from formal changes in Parliamentary oversight,
but from the goals of the particular majority that dominated Parliament at that time, so
that a different majority would have led to a different outcome, as in France. 22
These
examples suggest that to understand just how formal institutions work, we need to know
who uses them and for what purposes. 23
In the case of corporate governance, these considerations compel a look at the
interest group politics that surround policy toward MSP. Laws regulating corporations
and shareholders express the demands of groups acting through politics, whose support
politicians and bureaucrats need create and enforce legislation. Finance theorists begin
analysis of governance moral hazard issues by positing the goals, or interests, of investors
17
3/2/16 world bank , Gourevitch revisions
and managers: how, they ask, are investors, the principles, able to monitor the behavior
of the managers, the agents, given their principal’s great need of the expertise and
initiative of the agents. This helps model incentives inside the firm, but to explain public
policy, we need to examine how the goals of those inside the firm are translated into
public policy processes outside the firm. To that we need to think about interest groups
and their actions through political parties, elections, lobbying regulators and the familiar
trappings of political life.
We can specify key social actors by drawing on finance theory’s concern with
owners and managers. To this we add workers, an important group in society as well as in
the firm.
These can combine in various ways to generate alternative support (or
opposition) to laws that influence corporate governance outcomes, such as MSP. These
coalitional possibilities and realities can be shown graphically, as in Table 2.324.
18
3/2/16 world bank , Gourevitch revisions
Table 2 Political Coalitions and Governance Outcomes
Predicted
Coalitional Lineup
Winner
Model Name
Outcome
Investor
Diffusion
Labor
Blockholding
Social Bargain
Blockholding
Owners
Oligarchy
Blockholding
Owners + Workers
Transparency
Diffusion
Managers
Managerism
Diffusion
Pair A: Class Conflict
Owners + Managers vs.
Workers
Owners +
Managers
Owners + Managers vs.
Workers
Workers
Pair B: Sectoral
Owners vs. Managers +
Managers +
Workers
Workers
Owners vs. Managers +
Workers
Pair C: Property and
Voice
Owners + Workers vs.
Managers
Owners + Workers vs.
Managers
19
3/2/16 world bank , Gourevitch revisions
Each “pair” is based on a different force of political cleavage, separating society
into alternative policy views. The first pair pits workers against owners and managers.
Roe25 provides evidence that labor related groups oppose strong MSP, so that where they
are politically strong, we have weaker MSP. This mirrors the frequently made argument
that capital movement forces convergence around high MSP, as investors demand legal
protections for their stock purchasing.26 This model of political differences explains the
supply or absence of high MSP by the distribution of votes on a left-right continuum: the
bigger the labor vote (left) the weaker are MSP, and the bigger the conservative vote the
stronger are MSP.
The second alliance pair sees political divide along cross class, or sector patterns,
where workers, managers and inside owners support a mutually reinforcing regime of
blockholding. The coalition members prefer stability of existing arrangements of
employment, investment and leadership to the fluid relationships of the high MSP
systems. Minority shareholders are seen as outsiders, seeking to uproot stable
relationships built up over time. That stability sustains a distinctive pattern of production
which saves in transactions costs and specializes in high quality products and producer
goods. 27 Germany is the most well known examples of this corporate governance
pattern. It involves extensive collaboration among members of the system through a set
of arrangements frequently labeled as corporatist.
While both of these coalition patterns have received considerable attention in
literature on political economy, the third has not. The third coalition puts external
20
3/2/16 world bank , Gourevitch revisions
investors in alliance with workers against blockholders and managers in defense of high
levels of MSP and transparency generally. Pension funds are a key ingredient of this
coalition. The growth of pension funds may be causing an historical shift in the
coalitional patterns around public policy concerning MSP and corporate governance. 28
Reversing the expectation that that workers and investors are in conflict, we can see
motives that draw them together. Employee pension funds see MSPs as a way of
protecting an investment, and lobby parties dependent on labor support to get them. The
result is “left” governments doing more for investors than do the supposedly capital
friendly right wing governments which are more beholden to insiders and managers. 29
This pattern has been observed in Germany, where the SPD supported legislation that
strengthened shareholder protection and lowered the tax on capital gains from selling
shares, while the supposedly more business friendly CDU opposed these moves.30
The greater the share of country’s pension funds in private hands, rather than in
public Pay-Go systems, the higher is the level of MSP31, and the more likely we see
demands to strengthen them further. Just how actively pension funds use MSP powers
to impact corporate governance can be substantially affected by the micro-institutional
details of financial intermediaries. We know pension systems around the world have
troubles, in all countries rich and poor, highly developed and not. Much of the debate
about solutions has focused on the rate of return – do people do better with public or
private systems, taking into account the degree of risk? Of course the rate of return is
hugely important. But that issue, and impact of pension funds on corporate governance, is
connected to the micro details of the institutional investors which actually manage these
privatized funds: who controls those, are they in the hands of the same people who
21
3/2/16 world bank , Gourevitch revisions
control the firms where the money gets invested, what is the fee structure, how are these
Institutional Investors regulated, how much competition, what control to investors have
over the funds and their money, what powers to Institutional Investors have to challenge
management? These details influence both the rate of return and the impact of privatized
funds on corporate governance.32
What brings these coalitions together, causing some to win at some points, lose at
others? Institutionalists stress the role of formal structures in shaping the way clusters of
preferences aggregate: thus the argument that it is the electoral law or Majoritarian
structures which shapes the victory of different coalitional patterns among countries. The
de facto power approach looks instead at the resources various groups have to mobilize
for political action ( votes, organizations, the ability to strike). Variance among countries
in outputs expresses differences in the political strengths of these various interest groups.
Historical context , path dependence, and comparative political economy: The
interest group approach can help clarify the role of historical context and path
dependence in shaping outcomes. The demands of interest groups vary across countries
in part because of incentives shaped by prior choices in pubic policy, the results of
earlier conflicts. Rajan and Zingales’s33 description of the “great reversals” can be seen
as noting the impact of interest groups to economic dislocations after 1914: domestic
groups reacted against the disruptive effects of free trade and sought stability via
regulation. Among the effects was a shift toward bank influence and away from open
equity markets. Similarly, the “transplant effect” analyed by Berkowitz and Pistor34 can
be interpreted as the impact of interest groups within each country toward the costs, and
benefits, of adopting foreign practices. Roe35 explores the impact of war on interest group
22
3/2/16 world bank , Gourevitch revisions
alignments; countries which experienced considerable damage from the World Wars
shifted toward blockholding. Perrotti and von Thadden36 look at the impact of
hyperfinflation
during the 1920s: countries whose middle class investors were wiped
out shifted toward preferring the regulatory bank centered model. Rogoswki, Hiscox,
Frieden, Gourevitch examine the impact of world trading patterns on policy choices:
contractions in the economy produce demands for protection, while prosperity prepares
the ground for free trade. 37
These policy choices, made in response to trade or war or depression, generate
mutually reinforcing patterns of regulation and interest groups preferences. Hall and
Soskice 38 stylize market economies in two contrasting models, or production systems:
Liberal Market Economies (LMEs, which have fluid labor relations, vigorous anti -trust,
general skill -oriented educational systems, strong product market competition, and
diffuse shareholder oriented corporate and governance systems), and Coordinated
Market Economies (CMEs , which have highly structured labor relations, limited
domestic competition, price and market coordination, specific skill oriented educational
training, and blockholding governance systems.) Drawing on Milgrom and Roberts,39
they see strong “institutional complementarity”—placement on one dimension of the
economy reinforcements its complement on another. Thus investing in worker training
makes sense if there is low market fluidity in market share and employment and patient
capital shielded from shareholder pressures. Once systems of this kind start, at some
point in the past, the pieces reinforce each other. Management, blockholders and labor
generate reasons to cooperate around preservation of the firm. The interests and politics
of each production factor are thus strongly influenced by context. What labor,
23
3/2/16 world bank , Gourevitch revisions
management or capital want depends in part on which national production system they
operate in.
Institutional complementary leads to a criticism of the sufficiency of MSP levels
as an explanation of corporate governance patterns. So far, we have emphasized
alternative explanations of the provision of Minority Shareholder Protections-- legal
family, formal political institutions, and interest group politics. But the logic of
institutional complimentarity challenges the sufficiency of MSP more broadly. Investors
willingness to take minority positions is affected not only by the formal rules of MSP,
however these are generated, but by the rules that shape labor relations, product market
competition, price setting, and labor training. Roe notes where labor is strong, investors
may prefer blockholding to maintain stronger bargaining posture with labor, and labor in
turn may sustain militancy to deal with blockholding management. Power relationships
and policy regulation out side the strict domain of corporate governance as defined by
MSP thus have an impact on corporate governance patterns. This may explain the limited
power of correlating MSP with shareholder concentration: other variables are at play40 .
Another critique of the MSP approach moves away from regulation and law
altogether, Shareholder diffusion began, argue several specialists41, before regulations:
the listing requirements of the London and New York stock exchanges provided “private
bonding” reassurances. In the UK, mergers occurred among local elites, who knew each
other well. The demand for formal shareholder protections occurred later, as shareholding
spread. In addition to the historical evidence for this pattern, it is reinforced by a political
logic: who would lobby for shareholder protection if there were no shareholders? The
social foundations of markets can be quite important, worth more analysis. 42
24
3/2/16 world bank , Gourevitch revisions
Historical patterns, earlier choices, path dependence, and institutional
complimentarity provide a way of understanding the context of political choices
connected to corporate governance and economic policy outcomes. Policy choices at a
given moment reflect the interaction of interest group preferences, ideology and political
institutions. These reinforce each other – policy sustains the interests of those who put it
through. Along come shocks or changes in incentives: changes in the terms of trade, for
example between agriculture and industry, or recession; war and the development of
political obstacles to trade; economic depression, inflation. These change incentives and
alter political coalitions; they may change formal political institutions, creating a more
open political game. Policy may congeal around a new policy framework, which
reinforces itself again—till the next change in conditions or external shocks.
Conclusion: Political variables --- the interaction of preferences and institutions
to produce policy outputs – shape the way formal laws, rules and procedures work. An
account of law without a model of politics seems limited in predictive power and policy
utility. A strong model of politics is a difficult but vital goal. Several points can be
drawn from this discussion.
1. One size does not fit all.
will work the same way everywhere.
We can’t assume that the same formal institutions
We should be cautious therefore about telling
countries to adopt the same codes and laws on corporate governance, when they have
such greatly differing contexts with which to enforce and apply them. The best practice
codes on corporate governance call for strong MSP, but these rely on enforcement
mechanisms which lack support in much of the world. The diffuse shareholding model
takes more regulation, not less, and as the US Enron et. al scandals show, can be hard to
25
3/2/16 world bank , Gourevitch revisions
do even in highly developed legal systems. Blockholders , or various forms of private
bonding, may thereby do a better job under certain circumstances. We know a lot about
the ability of private bonding mechanisms to substitute for weak formal enforcement.43
These insights can be built into policy advice.
2. Formal rules and practice--- Don’t confuse formal rules with what happens.
It is commonly understood that formal law is not the same as practice. That laws
and codes are adopted does not mean we have actually changed the incentives social
actors have nor what they actually do. In the field of corporate governance, for example,
countries around the world have adopted good governance codes following
recommendations from OECD, and many other groups. Good codes are better than bad
ones. But codes by themselves are not practice. We cannot measure changes in
corporate governance activities by noting the adoption of these codes. We can see if
codes get translated into laws, but even then we cannot be sure the laws are enforced.
3. Enforcement requires a “politics” committed to those goals. Without political
support, good laws and codes are not enforced. Studies of the “transplant” effect
44
show this well: adopting codes does not produce the desired effect if the conditions for
the transplant -- a politics receptive to it --- are lacking.
4. Law is made by political process. Courts and lawyers operate in a context of
legislation and enforcement policy generated by political choices. The actual content of
law expresses political realities. Good or bad codes can be generated out of the same
intellectual roots.
26
3/2/16 world bank , Gourevitch revisions
5. Political processes are shaped by formal political institutions.
The de jure
forms of political process influence outputs, including the content of laws and
enforcement. De jure political processes trump legal formalism.
6 “ De facto “ power trumps “de jure” power. Informal mechanisms of social
power are able to adapt formal political institutions to their purposes. Thus de facto
power trumps de jure power which trumps legal formalism. It is surely the case that
these variables interact ---- law, formal institutions and social power all operate in a
complex system of interaction. But it is helpful to push at the endogeneity of that
interaction and tease out ways in which these interactions nestle.
This reasoning leads to a compelling research project: a World Bank Handbook of
de Facto Forms of Power. The important work on de jure institutions needs a counterpart
on social forms of power – on the distribution of landholding, on private armies and guns,
on clientalist networks, on cross-holding and overlapping patterns of ownership, on
interest group structures, on press concentration and control, on families45 and private
bonding mechanisms. This sort of material can help make progress in unraveling the
interaction of legal formalism, formal political institutions and the social distribution of
de facto power.
Several policy implications flow from these observations. MSP and diffuse
shareholding models require extensive law and regulations; the liberal market economy
rests on rules. Without the strong foundations of social support for these rules and their
enforcement, this model cannot work; blockholding may be a serious alternative.
Investing resources into legal training may not be effective if it is not combined with
analysis of political power balances. Investing in mechanisms of free speech, responsible
27
3/2/16 world bank , Gourevitch revisions
politics, economic journalists, trade union leaders, and business experts may be more
effective. Allowing variety of treatment across countries may be essential.
Thanks to both the Russell Sage Foundation where this paper was written on a fellowship
2005-06, and to the Guggenheim Memorial Fellowships for support. Thanks for
comments on earlier drafts to Stijn Claessens, Simon Johnson, James Shinn, and
Christopher Woodruff, and thanks to Lauren Porzelt for editorial assistance.
1
The problem of managers escaping the control of investors was famously observed by
A.A. Berle and Gardner Means, The Modern Corporation and Private Property (
McMillan 1933). While country specialists knew the US pattern was by no means
universal, La Porta et. al, helped develop recent awareness of the variance: R. La Porta ,
F. Lopez-de-Silanes, A. Shleifer & R. W. Vishny, Legal Determinants of External
Finance, 52 The Journal of Finance 1131 (1997); La Porta, et. al. , Law and Finance,
106 Journal of Political Economy 1113 (1998); La Porta, et. al , Investor Protection and
Corporate Governance 58 Journal of Financial Economics 3 ( 2000); La Porta, et. al ,
Investor Protection and Corporate Valuation. 57 Journal of Finance 1147 (2002); R. La
Porta, F. Lopez-de-Silanes & A. Shleifer, Corporate Ownership Around the World, 54
The Journal of Finance 471 (1999). For an extensive review of the literature see Marco
Becht, Patrick Bolton & Ailsa Röell, Corporate Governance and Control, in Handbook
of the Economics of Finance, (Elsivier, 2005).
2
This particular use of the terms de jure and de facto derive from Daron Acemoglu &
Simon Johnson, Unbundling Institutions, 113 Journal of Political Economy 949 (2005).
See also Daron Acemoglou & James Robinson, The Persistence of Power, Elites and
Institutions, Mimeo (2006); Daron Acemoglou, Simon Johnson, James Robinson,
Institutions as the Fundamental Cause of Long-Run Growth (2005), in Handbook of
Economic Growth, vol. 1A, (North-Holland 2005); Daron Acemoglu, Simon Johnson &
James A. Robinson, The Colonial Origins of Comparative Development: An Empirical
Investigation, 91 American Economic Review 1369 (2001).
Daron Acemoglu, Constitutions, Politics and Economics; A Review Essay on Persson
and Tabellini’s "The Economic Effects of Constitutions", 43 Journal of Economic
Literature 1025 (2005).
28
3/2/16 world bank , Gourevitch revisions
3
Earlier work looked at bank debt vs. equity ratios. This is still the dominant form for
comparing patterns in earlier periods, as shareholder concentration is hard to document.
See Raghuram Rajan & Luigi Zingales, The Great Reversals: The Politics of
Financial Development in the 20th Century, 69 Journal of Financial Economics 5 (2003).
4
Japan ranks high on this list because the system of cross shareholding renders
meaningless the diffusion pattern of ownership—there is no effective market for control
there. China ranks high because the stock market captures only a small portion of its
economy, most of which remains in state hands.
5
This passage draws on Peter Gourevitch & James Shinn, Political Power and Corporate
Control: The New Global Politics of Corporate Governance (Princeton University Press
2005), especially chapters two and three.
6
Historical research questions the causal sequence: do shareholders precede MSP or does
MSP come ahead of shareholder. Private bonding arrangements via stock exchanges or
mergers, may have created a group of shareholders who then demanded MSPs. That
makes political sense (why create something for which there is no constituency), and may
be historically accurate, but it is not clear how it handles the present: can formal MSP
regulations and their enforcement induce shareholding. See Brian Cheffins, Does Law
Matter?: The Separation of Ownership and Control in the United Kingdom, 30 Journal
of Legal Studies 459 (2001); Brian Cheffins, Corporate Law and Ownership Structure:
A Darwinian Link?, 25 University of New South Wales Law Journal 346 (2002);
Brian Cheffins, Putting Britain on the Roe Map: The Emergence of the Berle-Means
Corporation in the United Kingdom, in Corporate Governance Regimes: Convergence
and Diversity, 147-174.(Oxford University Press 2002 ); Randall Morck, ed., History of
Corporate Ownership, (University of Chicago Press 2006); John Coffee, The Rise of
Dispersed Ownership: The Roles of Law and the State in the Separation of Ownership
and Control, 111 The Yale Law Journal 1 (2001).
7
The ranks are a composite constructed by the authors from numerous sources. See
Gourevitch & Shinn, supra note 5, p. 48, the sources for Table 3.1.
8
Gourevitch & Shinn, supra note 5, p 51, and Table A-4, p. 301. The correlation is
strongest for developed countries, at –.42 ** (p<.05) at significance of .03 at, while
reversed in sign and weak significance for developing countries (.29).
9
Raghuram Rajan & Luigi Zingales, The Great Reversals: The Politics of Financial
Development in the 20th Century, 69 Journal of Financial Economics 5 (2003).
10
Mark Roe, Strong Managers, Weak Owners: The Political Roots of American
Corporate Finance (Princeton University Press 1994).
29
3/2/16 world bank , Gourevitch revisions
11
Daniel Berkowitz, Katharina Pistor & Jean-Francois Richard, Economic development,
legality and the transplant effect 47 European Economic Review 165 (2003).
12
Daron Acemoglu & Simon Johnson, supra note 2; Christopher Woodruff, Measuring
Institutions (2005) in Handbook on Corruption (forthcoming); Luc Laeven &
Christopher Woodruff, The Quality of the Legal System: Firm Ownership and Firm Size,
Mimeo (2005).
13
Glaesar et.al make important warning about the need to be sure measures of institutions
are independent from those of the outputs they produce. E. Glaeser, R. La Porta, F.
Lopez-de-Silanes, A. Shleifer, Do Institutions Cause Growth?, 9 Journal of Economic
Growth 271 (2004).
14
See Acemolgu, Johnson & Robinson, supra note 2; C. Douglas North, Structure and
Change in Economic History (W.W. Horton 1981); S. Haber, A. Razo, and N. Maurer,
The Politics of Property Rights: Political Instability, Credible Commitments, and
Economic Growth in Mexico 1876-1929 (Cambridge University Press 2003).
15
Thorsten Beck, George Clarke, Alberto Groff, Philip Keefer & Patrick Walsh, New
Tools in Comparative Political Economy: The Database of Political Institutions 15
World Bank Economic Review 165 (2001). Arend Lijphart, Patterns of Democracy:
Governmnent Forms and Performance in Thirty-Six Countries (Yale University Press
1999); Torsten Persson & Guido Tabellini, The Economic Effects of Constitutions: What
do the Data Say? (MIT Press 2003); Matthew S. Shugart & John M. Carey, Presidents
and Assemblies: Constitutional Design and Electoral Dynamics (Cambridge University
Press 1992); Stephan Haggard & Mathew C. McCubbins, Presidents, Parliament and
Policy (Cambridge University Press 2001).
16
Peter Gourevitch & Michael B. Hawes, The Politics of Choice among National
Production Systems, L’Année de la regulation, Presses de Sciences Po 2002) ;
Gourevitch & Shinn, supra, note 5.
17
Peter Hall and David Soskice , Varieties of Capitalism: The Institutional Foundations
of Comparative Foundation , (Oxford University Press, 2001); Torben Iversen & David
Soskice, Electoral Institutions, Parties, and the Politics of Class: Why Some
Democracies Redistribute More than Others, 96 American Political Science Review,
forthcoming (2006); Torben Iversen & David Soskice, An Asset Theory of Social Policy
Preferences , 95 American Political Science Review 875 (2001); Isabela Mares, The
Politics of Social Risk (Cambridge University Press 2003).
18
Marco Pagano & Paolo Volpin, The Political Economy of Corporate Governance, 95
American Economic Review 1005 (2005). Marco Pagano and Paolo Volpin, Workers,
Managers, and Corporate Control 60 The Journal of Finance 841(2005); Marco
30
3/2/16 world bank , Gourevitch revisions
Pagano & Paolo Volpin, The Political Economy of Finance, 17 Oxford Review of
Economic Policy 502 (2001).
19
Gourevitch & Shinn, supra note 5; Raymond Fisman, Estimating the Value of
Political Connections, 91 American Economic Review 1095 (2001); Andrew McIntyre,
Institutions and Investors: The Politics of the Financial Crisis
in Southeast Asia, 55 International Organization 81(2001).
20
Acemoglou and Johnson, supra, note 2.
21
A similar concern with social forms of power can be found in Raghuram Rajan & Luigi
Zingales, Creating Constituencies for Reform, Mimeo (2005).
22
David Stasavage, Credible Commitment in Early Modern Europe: North and Weingast
Revisited, 18 Journal of Law, Economics, and Organization 155 (2002).
23
Rajan & Zingales, supra note 21.
24
Gourevitch & Shinn, supra note 4.
25
Mark Roe, Political Determinants of Corporate Governance (Oxford University Press,
2003).
26
Paul MacAvoy & Ira Milstein, The Recurrent Crisis on Corporate Governance
(Palgrave 2003); Henry Hansmann & Reiner Kraakman, The End of History for
Corporate Law (Harvard University Press 2000).
27
W. Carl Kester, American and Japanese Corporate Governance: Converging to Best
Practice? In National Diversity and Global Capitalism, Suzanne Berger and Ronald
Dore, es, (Cornell University Press, 1996) ; Peter Hall and David Soskice , supra , note
17
28
Predicted by Peter Drucker, The Unseen Revolution: How Pension Fund Socialism
Came to America (Harper Collins 1976).
29
Hoepner Martin, European Corporate Governance Reform and the German Party
Paradox, Max-Planck-Institute for the Study of Societies Program for the Study of
Germany and Europe Working Paper Series 03.1 (2003); John W. Cioffi & Martin
Hoepner, The Political Paradox of Corporate Governance
Reform: Why the Center-Left is the Driving Force Behind the Rise of Financial
Capitalism. Paper presented at the 2004 Annual Meeting of the American Political
Science Association, September 2-5, in Chicago, Illinois.
.
30
Hoepner & Cioffi, supra, note 29.
31
31
3/2/16 world bank , Gourevitch revisions
See Gourevitch & Shinn, supra note 5.
32
See Gourevitch & Shinn, supra note 5; Peter Gourevitch & Jacob Allen, Chilean
Pension Reform and Corporate Governance, Graduate School of International Relations
and Pacific Studies, UCSD, Mimeo, (2006); Fernando Lefort & Eduardo Walker,
Ownership and Capital Structure of Chilean Conglomerates: Facts and Hypotheses for
Governance. 3 Revista ABANTE 3 (1999).
35
33
Rajan and Zingales, supra , note 3.
34
Berkowitz and Pister, supra, note 11.
Mark Roe, Legal Origins and Stock Markets in the Twentieth Century, Mimeo (2006).
36
Enrico Perotti & Ernst-Ludwig von Thadden, The Political Economy of Corporate
Control and Labor Rents, 114 Journal of Political Economy 145 –174 (2006); Enrico
Perotti and Ernst-Ludwig von Thadden, Dominant Investors and Strategic
Transparency, 21 Journal of Law and Economics 76 ( 2005) ; Enrico Perotti and
Ernst-Ludwig von Thadden, Corporate Governance and the Distribution of
Wealth, 162 Journal of Institutional and Theoretical Economics 204 (2006).
37
Ronald Rogoswki, Commerce and Coalitions (Princeton 1989); Peter Gourevitch,
Politics in Hard Times, (Cornell 1986); Michael Hiscox, International Trade and
Political Conflict: Commerce, Coalitions, and Mobility, (Princeton University Press
2002); Jeffry Frieden, Global Capitalism (Norton 2006).
38
Hall & Soskice, supra, note 17.
39
Paul Milgrom & John Roberts, The Economics of Modern Manufacturing:
Technology, Strategy and Organization, 80 American Economic Review 511 (1990).
40
See Gourevitch & Shinn, supra, note 5; Hall & Soskice, supra, note 17 ; and Mark
Roe, supra, note 10.
41
See Cheffins, supra note 6; Julian Franks, Colin Mayer & S. Rossi, Spending Less
Time with the Family: The Decline of Family Ownership in the UK, in The History of
Corporate Governance Around the World: Family Business Groups to Professional
Managers (University of Chicago Press 2005); Julian Franks, Colin Mayer & Hannes
Wagner, The Origins of the German Corporation – Finance, Ownership and Control,
Mimeo, unpublished, (2005). See Randall Morck, Daniel Wolfenzon & Bernard Yeung,
Corporate Governance, Economic Entrenchment and Growth, 43 Journal of Economic
Literature 655 (2005).
32
3/2/16 world bank , Gourevitch revisions
42
Avner Greif, Paul Milgrom & Barry Weingast, Coordination, Commitment and
Enforcement: The Case of the Merchant Gild, 102 The Journal of Political Economy 745
(1994); Avner Greif, Contract Enforceability and Economic Institutions in Early Trade:
The Maghribi Traders' Coalition, 83 American Economic Review 525 (1993). For an
evaluation of how decision-maker understanding of their options shapes behavior, see
Pepper D. Culpepper, Institutional Change in Contemporary Capitalism: Coordinated
Finacial Systems since 1990, 57 World Politics 173 (2005).
43
44
45
John McMillan & Christopher Woodruff, The Central Role of Entrepreneurs in
Transition Economies,16 Journal of Economic Perspectives 153 (2002); Simon
Johnson, John McMillan & Christopher Woodruff, Property Rights and Finance,
92 American Economic Review 1335 (2002).
Daniel Berkowitz, Katharina Pistor & Jean-Francois Richard, supra, note 11.
See for example Marianne Bertran, Simon Johnson, Krislert Samphanthrarak &
Antoinette Schoar, Mixing Family with Business: A Study of Thai Business Groups and
the Families Behind Them, Mimeo (2004); Stephen Haber, Noel Maurer, and Armando
Razo, When Law Does Not Matter: The Rise and Decline of the Mexican Oil Industry, 63
Journal of Economic History 1 (2003); Alfred Stepan, Rethinking Military Politics:
Brazil and the Southern Cone (Princeton University Press 1988).