Legal Origins: Reconciling Law & Finance
and Comparative Law
Mathias M. Siems*
Law and finance scholars have increasingly relied
upon comparative law in the last few years. The work
of these scholars has considered, in empirical terms, the
effect that legal rules and their enforcement have on
financial development in different countries. These
traditional
studies have
distinction between civil
law
countries. Whether this revival of legal families (or
legal origins) is a useful way forward is, however, a
matter of debate. The author challenges
the
methodology these studies adopt and argues instead for
reliance on characteristic features of national legal
systems, as distinct from systemic origins, as a basis for
analysis.
law and common
routinely adopted
the
Les auteurs oeuvrant dans le domaine du droit et
de la finance font de plus en plus recours au droit
comparatif. Les recherches acadmiques se situant
lintersection du droit et de la finance arborent une
tendance qui se veut de vouloir quantifier les effets
quont les rgles de droit ainsi que limpact de leurs
mise en application sur le dveloppement financier de
divers pays. De plus, les donnes manant de ces tudes
sont souvent attribues et lies la distinction
traditionnelle qui spare les pays de tradition civiliste et
ceux issus de la common law. La question quant
savoir si le renouveau de ces familles lgales (ou
origines
lgales) constitue une avance nest
cependant pas sans controverse. Lauteur remet en
cause ces recherches et prconise une approche plus
prcise et significative qui consiste en une analyse
oriente sur les traits spcifiques et non sur le concept
de familles lgales.
* Reader, School of Law, University of Edinburgh; Research Associate, Centre for Business
Research, University of Cambridge; Senior Research Associate, Peterhouse, University of Cambridge.
I am grateful to the comments of Oscar Alvarez Macotela, John Armour, Simon Deakin, Sonja
Fagernas, Martin Gelter, Priya Lele, Holger Spamann, Dirk Zetzsche, and two anonymous reviewers.
Mathias M. Siems 2007
To be cited as: (2007) 52 McGill L.J. 55
Mode de rfrence : (2007) 52 R.D. McGill 55
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
56
Introduction
I. Legal Origins in the Law and Finance Literature
A. The Historical Origins
B. Comparative Statistical Analysis
II. Legal Families: A Problematic Distinction
III. Unbundling Legal Families: The Search for
Characteristic Features
A. Unreliable Categories
B. Four Criteria for Effective Differentiation
IV. Re-examining the Differences
A. An Example from the Legal Origins Literature
B. An Alternative Methodology
1. Methodology
2. Explaining the Categories
3. The Results
Conclusion
[Vol. 52
57
58
58
60
62
70
70
72
73
74
75
78
78
79
80
2007] M.M. SIEMS RECONCILING LAW & FINANCE AND COMPARATIVE LAW
57
[T]he comparative lawyer cannot restrict his field …
narrowly. More than any other academic[,] … he
must be prepared to find new topics for discussion
and research.1
Introduction
It has been said that the twenty-first century will be an era of comparative law.2
The problem, however, is how properly to engage in comparative law. Politicians
and judges allegedly pay no attention to comparative law because it is regarded as too
complicated and theoretical for a generalist audience.3 Recently, however, law and
finance scholars have discovered the usefulness of comparative law.4 Their studies
have had an immense impact in academic fields such as comparative corporate
governance,5 and their findings are given great weight by the World Bank when
assessing the quality of law and legal institutions in a particular state.6 In substance,
law and finance scholars look at the quantifiable effect that legal rules and their
enforcement have on financial development. They ask whether specific legal features,
such as a particular legal rule or the effectiveness of courts, correlate with financial
data, such as a countrys stock market capitalization. Another pervasive feature of
these studies is their reliance on the traditional distinction between common law and
civil law countries as an analytical tool. This revival of legal familiesor legal
origins as these studies call it7may surprise modern comparative lawyers.
1 F.H. Lawson, The Field of Comparative Law (1949) 61 Jurid. Rev. 16 at 36.
2 Esin rc, The Enigma of Comparative Law (Leiden, Neth.: Martinus Nijhoff, 2004) at 216.
3 See Basil Markesinis, Comparative Law in the Courtroom and Classroom: The Story of the Last
Thirty-five Years (Oxford: Hart, 2003) at 61-62. See also Basil Markesinis, Comparative LawA
Subject in Search of an Audience (1990) 53 Mod. L. Rev. 1; Basil Markesinis, Foreign Law and
Comparative Methodology (Oxford: Hart, 1997) at 3.
4 See infra note 10ff.
5 For a recent overview, see Cally Jordan, The Conundrum of Corporate Governance (2005) 30
Brook. J. Intl L. 983.
6 See The World Bank Group, Doing Business: Benchmarking Business Regulations, online: Doing
Business
Starting a Business, Employing Workers, Getting Credit, Protecting Investors, and
Enforcing Contracts). For critical comments, see Association Henri Capitant des Amis de la Culture
Juridique Franaise, Les droits de tradition civiliste en question propos des rapports Doing
Business de la Banque Mondiale, vol. 2 (Paris: Socit de Lgislation Compare, 2006); Claude
Mnard & Bertrand du Marais, Can We Rank Legal Systems According to Their Economic
Efficiency? in Peter Nobel, ed., New Frontiers of Law and Economics (St. Gallen, Switz.: Schulthess,
2006); Mathias M. Siems, Statistische Rechtsvergleichung Rabels Zeitschrift fr auslndisches und
internationales Privatrecht [forthcoming].
7 The terminology employed in the field of comparative law overlaps to a considerable degree:
legal origins, legal families, and legal traditions are often employed interchangeably. In this
article, the term legal origin is reserved for instances when I am referring specifically to studies that
that
law
[Vol. 52
is becoming
increasingly emphasize
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
58
Comparatists
international,
transnational, or even global, such that looking at legal traditions is considered less
important.8 In light of these different approaches, this article discusses the
appropriateness of the continued use made of legal families as analytical categories
by law and finance scholars. Part I summarizes the reasoning and results of the
studies conducted by law and finance scholars while Part II challenges these studies
by identifying problems in the distinction made between legal families. Part III
isolates characteristic features that allow for more precise analysis than the use of
legal families as such, and Part IV re-examines the differences between countries
through the lens of these new identifiers. The article concludes with an invitation to
incorporate these lessons from comparative law by employing new criteria in future
law and finance studies.
I. Legal Origins in the Law and Finance Literature
Statements by law and finance scholars on legal origins have two dimensions. On
the one hand, they describe the origins of different legal traditions and explain how
they have spread throughout the world. On the other, they deploy legal traditions as
categories to analyze whether legal rules foster economic development. This analysis
is performed quantitatively, by means of statistical regressions.9
A. The Historical Origins
Law and finance scholars trace the origins of different legal families to twelfth-
and thirteenth-century England and France.10 Both countries were faced with the
adopt that language. Otherwise, I will use legal family when referring to the analytical category, and
legal tradition when referring to the character of a legal order.
8 See Mathias Reimann, Beyond National Systems: A Comparative Law for the International Age
(2001) 75 Tul. L. Rev. 1103 at 1115; Esin rc, Family Trees for Legal Systems: Towards a
Contemporary Approach in Mark Van Hoecke, ed., Epistemology and Methodology of Comparative
Law (Oxford: Hart, 2004) 359 at 361 [rc, Family Trees]; Jaakko Husa, Classification of Legal
Families Today. Is it Time for a Memorial Hymn? [2004] R.I.D.C. 11; James Gordley, Common law
und civil law: eine berholte Unterscheidung (1993) Z. Eu. P. 498. See also Stefan Vogenauer, An
Empire of Light? Learning and Lawmaking in the History of German Law (2005) 64 Cambridge L.J.
481 (the oft-evoked distinction between the Common law and the Civil law does not prove to be
of much help at 483).
9 See e.g. Rafael La Porta, Florencio Lopez-de-Silanes & Andrei Shleifer, What Works in
Securities Law? (2006) 61 Journal of Finance 1 at 27 [La Porta, Lopez-de-Silanes & Shleifer,
Securities Law]. Statistical details are outside the scope of this article. In particular, the use of legal
origins as a supposed remedy to the problem of endogeneity will not be addressed.
10 For what follows, see Edward L. Glaeser & Andrei Shleifer, Legal Origins (2002) 117 The
Quarterly Journal of Economics 1193; Simeon Djankov et al., The New Comparative Economics
(2003) 31 Journal of Comparative Economics 595 at 605-606. Compare Thorsten Beck & Ross
Levine, Legal Institutions and Financial Development in Claude Mnard & Mary M. Shirley, eds.,
Handbook of New Institutional Economics (Dordrecht, Neth.: Springer, 2005) 251 at 254-58 (tracing
origins to the fifteenth century for France and to the sixteenth and seventeenth centuries for England).
59
2007] M.M. SIEMS RECONCILING LAW & FINANCE AND COMPARATIVE LAW
problem of how to protect law enforcers (such as judges) from coercion by litigants
through either violence or bribes. According to law and finance scholars, the crucial
difference between the two nations at the time was that France was less peaceful than
England. In France there was therefore a greater need for protection and control of
law enforcers by the state. Consequently, following the Roman law tradition, France
adopted a civil law system characterized by fact-finding by state-employed judges,
automatic review of decisions, and, later, a reliance on codes rather than judicial
discretion. In contrast, England developed a common law system that relied on fact-
finding by juries, independent judges, infrequent appeals, and judge-made law rather
than strict codes.
According to the law and finance literature, the English and French legal
traditions spread throughout the world through conquest, colonization, and
imitation.11 Apart from the French model, there are said to be two other civil law
traditions: the German and the Scandinavian. The seminal moment for the German
legal tradition is the adoption of the German Civil Code in 1900. Much like its French
counterpart, the German legal tradition is based on Roman civil law and was
subsequently exported to other countries.12 By contrast, the Scandinavian legal
tradition, which developed relatively independently in the seventeenth and eighteenth
centuries, is less closely linked with Roman civil law13 and has not spread throughout
the world.14 Finally, some studies refer to a socialist-transition legal family, which is
based on the legal tradition that emerged from the breakup of the Soviet Union.15
How exactly legal traditions spread and why a particular country belongs to a
particular family is usually not explained in detail. Rather, studies often refer
generally to mainstream comparative law books.16 Insofar as explanations are given,
they are usually very short; the act of categorizing a particular country appears
unproblematic. For example, it is said that Austrian and Swiss civil codes were
developed at the same time as the German civil code and the three influenced each
For a critical comment, see Association Henri Capitant des Amis de la Culture Juridique Franaise,
supra note 6 at para. 20.
11 See Beck & Levine, ibid. at 258-60.
12 See ibid. at 256, 258-59.
13 Some studies regard the Scandinavian countries as part of the civil law tradition. See e.g. Rafael
La Porta et al., Law and Finance (1998) 106 Journal of Political Economy 1113 at 1115 [La Porta et
al., Law and Finance]. Others treat it as a separate legal family. See e.g. La Porta, Lopez-de-Silanes
& Shleifer, Securities Law, supra note 9 at 14.
14 See Beck & Levine, supra note 10 at 257.
15 See Simeon Djankov, Caralee McLiesh & Andrei Shleifer, Private Credit in 129 Countries
Journal of Financial Economics [forthcoming in 2007], online: National Bureau of Economic
Research
16 See e.g. Ren David & John E. C. Brierley, Major Legal Systems in the World Today, 3d ed.
(London: Stevens & Sons, 1985); Konrad Zweigert & Hein Ktz, Introduction to Comparative Law,
3d ed. (Oxford: Clarendon Press, 1998); Thomas H. Reynolds & Arturo A. Flores, Foreign Law:
Current Sources of Cases and Basic Legislation in Jurisdictions of the World (Littleton, Colo.: Fred
Rothman & Company, 1989).
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
60
other heavily. In turn, Czechoslovakia, Hungary, Yugoslavia, and Greece relied on
German civil law in formulating and modernizing their legal systems in the early part
of the 20th century.17
[Vol. 52
B. Comparative Statistical Analysis
The question of whether legal rules foster economic development lies at the core
of law and finance studies and extends to the comparative analysis of legal traditions.
For instance, in one of their first studies, La Porta et al. found that the protection of
investors (i.e., shareholders and creditors) is not only strongly related to financial
development, but also that differences can be explained by reference to legal origin.18
Having examined forty-nine countries, the result of their calculations was that
common law countries protect shareholders and creditors better than civil law
countries (especially the French ones) do.19 Recently, Djankov, McLiesh & Shleifer
extended this study to the protection of creditors in 129 countries.20 In contrast to La
Porta et al., they looked not only at creditor protection by means of ex post private
dispute resolution, but also at the use of public credit registries. Because common law
countries rely primarily on the former ex post mechanism whereas French civil law
countries sustain their debt market through the latter ex ante institutions, this study
also identified differences according to legal origin.21 With respect to investor
protection, a study by La Porta, Lopez-de-Silanes & Shleifer on securities law in
forty-nine countries found that there is little evidence that public enforcement is
important, but that strong evidence exists to suggest that laws mandating disclosure
and facilitating private enforcement through liability rules benefit stock markets. This
result was also linked to the common law / civil law distinction. The common law
was said to emphasize market discipline and private litigation. The study found that
private monitoring and contracting were more important than public enforcement of
securities law, and the authors concluded accordingly that these advantages of the
17 Beck & Levine, supra note 10 at 258. In reality, however, the Austrian Civil Code (Allgemeines
Brgerliches Gesetzbuch) was enacted in 1811, well ahead of the 1900 German Civil Code
(Brgerliches Gesetzbuch). Furthermore, the Swiss Civil Code (Zivilgesetzbuch) was also heavily
influenced by the French Code civil. See Peter Tuor, Berhard Schnyder & Jrg Schmid, Einleitung
in Peter Tuor, ed., Das Schweizerische Zivilgesetzbuch, 11th ed. (Zurich: Schulthess, 1995) at 1.
18 La Porta et al., Law and Finance, supra note 13.
19 Ibid. at 1139. For a recent study focusing on self-dealing, which slightly modifies the shareholder
index, see Simeon Djankov et al., The Law and Economics of Self-Dealing (Working Paper No.
11883, National Bureau of Economic Research, 2005), online: National Bureau of Economic
Research
Mathias M. Siems, Shareholder Protection: A Leximetric Approach (2007) 7 Journal of Corporate
Law Studies 17.
20 Supra note 15.
21 Ibid. at 22.
61
2007] M.M. SIEMS RECONCILING LAW & FINANCE AND COMPARATIVE LAW
common law tradition were decisive for the superior quality of securities laws of
these countries.22
Other studies have confirmed that civil law countries put a stronger emphasis on
state involvement. For example, studies by La Porta and others found that in countries
of French legal origin there is more government ownership of banks23 and that state-
owned enterprises play a greater role than they do in common law countries.24
Djankov et al. analyzed the level of regulation in cases of entry by new firms in
eighty-five countries. The study found that (French) civilian countries regulate entry
more heavily than common law countries do.25 Furthermore, a study by Botero et al.
on labour markets in eighty-five countries found that social control of business by
labour law is higher in (French) civil law than in common law countries.26
Finally, law and finance scholars have identified two possible mechanisms
through which legal origin influences financial development.27 First, the relevance of
legal origins may be explained by the political channel, whereby the relative
priority given to private property rights and rights of the state is considered
important.28 As a proxy for this criterion, two studies examined judicial control of
political decisions. The studies found that, for instance, there is less supreme court
power in civil law countries than in common law countries.29 State influence was
inferred to be greater in the civil law tradition. Second, it could be decisive that legal
systems in different traditions respond differently to changing socio-economic
circumstances. With respect to this adaptability channel, the case law approach by
the common law is seen as more flexible whereas civil law is regarded as inherently
more formalistic and rigid.30 A study by Djankov et al. on legal proceedings in 109
22 La Porta, Lopez-de-Silanes & Shleifer, Securities Law, supra note 9 at 28. For a different view,
see Mark J. Roe, Legal Origins, Politics, and Modern Stock Markets (2006) 120 Harv. L. Rev. 460
at 491-94 [Roe, Legal Origins].
23 See Rafael La Porta, Florencio Lopez-de-Silanes & Andrei Shleifer, Government Ownership of
Banks (2002) 57 Journal of Finance 265 at 271.
24 Rafael La Porta et al., The Quality of Government (1999) 15 J.L. Econ. & Org. 222 at 261.
25 Simeon Djankov et al., The Regulation of Entry (2002) 117 The Quarterly Journal of
Economics 1 at 271.
26 Juan C. Botero et al., The Regulation of Labor (2004) 119 The Quarterly Journal of Economics
1339 at 1365. For a comment, see Simon Deakin & Beth Ahlering, Labor Regulation, Corporate
Governance and Legal Origin: A Case of Institutional Complementarity? Law & Socy Rev.
[forthcoming in 2007].
27 For a good summary, see Beck & Levine, supra note 10 at 271-73.
28 Thorsten Beck, Asli Demirg-Kunt & Ross Levine, Law and Finance: Why Does Legal Origin
Matter? (2003) 31 Journal of Comparative Economics 653 at 654.
29 See ibid. at 667. For other criteria such as the likelihood of judges receiving tenure and judicial
review of the constitutionality of laws, see Rafael La Porta et al., Judicial Checks and Balances
(2004) 112 Journal of Political Economy 445.
30 Beck & Levine, supra note 10 at 261-62. For a more comprehensive view of legal adaptability,
see Mathias M. Siems, Legal Adaptability in Elbonia (2006) 2 International Journal of Law in
Context 393.
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
62
countries confirms this result. Djankov et al. looked at numerous variables, such as
the need for legal representation and written documents, statutory justification for a
complaint, and limits on the use of evidence. They found that civil law countries (in
particular the francophone ones) provide a more formalist approach than do common
law countries.31
[Vol. 52
II. Legal Families: A Problematic Distinction
Law and finance studies have received a lot of criticism.32 In particular, the
relevance of legal origins for financial development is doubtful. There are scholars
who emphasize that other aspects of a society, such as politics,33 culture or religion,34
and geographical institutions35 are considerably more determinative than belonging to
a particular legal tradition. If one disregarded these other social elements, one could
easily come to absurd results. This was made very clear in a paper by West, which
examines whether legal origin influences success in the World Cup of soccer!36 He
31 Simeon Djankov et al., Courts (2003) 118 The Quarterly Journal of Economics 453.
32 See supra note 6. On the choice of variables and their coding, see Lele & Siems, supra note 19;
Mathias M. Siems, What Does Not Work in Comparing Securities Laws: A Critique on La Porta et
al.s Methodology (2006) 16 International Company & Commercial Law Review 300 [Siems, A
Critique on La Porta et al.]; Sofie Cools, The Real Difference in Corporate Law Between the United
States and Continental Europe: Distribution of Powers (2005) 30 Del. J. Corp. L. 697; Udo C.
Braendle, Shareholder Protection in the USA and Germany: Law and Finance Revisited (2006) 7
German Law Journal 257; Markus Berndt, Global Differences in Corporate Governance Systems
(Wiesboden, F.R.G.: Gabler Edition Wissenschaft, 2002) at 17-18; Holger Spamann, On the
Insignificance and/or Endogeneity of La Porta et al.s Anti-Director Rights Index under Consistent
Coding (Discussion Paper No. 7, Harvard Law School John M. Olin Center for Law, Economics, and
Business, 2006), online: Harvard Law School John M. Olin Center for Law, Economics, and Business
33 See Mark J. Roe, Political Determinants of Corporate Governance (Oxford: Oxford University
Press, 2003); Roe, Legal Origins, supra note 22; Marco Pagano & Paolo Volpin, The Political
Economy of Finance (2001) 17 Oxford Review of Economic Policy 502; Marco Pagano & Paolo
Volpin, The Political Economy of Corporate Governance (2005) 95 American Economic Review
1005; Peter A. Gourevitch & J. James Shinn, Political Power and Corporate Control: The New
Global Politics of Corporate Governance (Princeton, N.J.: Princeton University Press, 2005).
34 See Ren M. Stulz & Rohan Williamson, Culture, Openness, and Finance (2003) 70 Journal of
Financial Economics 313; Amir N. Licht, Chanan Goldschmidt & Shalom H. Schwartz, Culture,
Law, and Corporate Governance (2005) 25 Intl Rev. L. & Econ. 229.
35 See Daron Acemoglu, Simon Johnson & James A. Robinson, The Colonial Origins of
Comparative Development: An Empirical Investigation (2001) 91 American Economic Review
1369.
36 Mark West, Legal Determinants of World Cup Success (Discussion Paper No. 009, University
of Michigan John M. Olin Center for Law & Economics, 2002), online: University of Michigan Law
School
2007] M.M. SIEMS RECONCILING LAW & FINANCE AND COMPARATIVE LAW
found that a legal order influenced by the French civilian tradition correlated with
World Cup success in a statistically significant way:
63
Perhaps teams from countries with systems based on the French model (such as
1998 champion France and 2002 champion Brazil) perform well due to the
remaining vestiges of the Napoleonic Code that somehow remove discretion
from coaches and managers in the same manner that the civil law system
curtails judicial activism. Or maybejust maybesome other forces are at
work.37
The search for other factors is indeed the usual reaction to strange statistical results,
as robustness checks may expose a bias due to a limited breadth of variables
considered.38
However, the main problem of legal traditions and law and finance lies
somewhere else. Before considering different channels and checking the robustness
of the results, one must first determine which legal system belongs to which legal
family. For instance, in the Djankov, McLiesh & Shleifer study on creditor rights, the
classification appears as follows:
37 West, ibid. at 5.
38 See Beck & Levine, supra note 10 at 268-69.
64
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 52
Legal Origin39
English
Legal Origin
French
Legal Origin
German
Legal Origin
Nordic
Legal Origin
Socialist
Legal Origin
Countries
Australia, Bangladesh, Botswana, Canada,
Ethiopia, Ghana, Hong Kong, India, Iran,
Ireland, Israel, Jamaica, Kenya, Lesotho,
Malawi, Malaysia, Namibia, Nepal, New
Zealand, Nigeria, Pakistan, Papua New
Guinea, Saudi Arabia, Sierra Leone,
Singapore, South Africa, Tanzania, Thailand,
Uganda, United Arab Emirates, United
Kingdom, United States, Yemen, Zambia,
Zimbabwe.
Albania, Algeria, Angola, Argentina, Belgium,
Benin, Bolivia, Brazil, Burkina Faso, Burundi,
Cambodia, Cameroon, Central African
Republic, Chad, Chile, Colombia, Democratic
Republic of the Congo, Republic of the Congo,
Costa Rica, Cote dIvoire, Dominican
Republic, Ecuador, Egypt, El Salvador,
France, Greece, Guatemala, Guinea, Haiti,
Honduras, Indonesia, Italy, Jordan, Kuwait,
Lao PDR, Lebanon, Lithuania, Madagascar,
Mali, Mauritania, Mexico, Morocco,
Mozambique, Netherlands, Nicaragua, Niger,
Oman, Panama, Paraguay, Peru, Philippines,
Portugal, Puerto Rico, Romania, Rwanda,
Senegal, Spain, Syria, Togo, Tunisia, Turkey,
Uruguay, Venezuela, Vietnam.
Austria, Bosnia and Herzegovina, Bulgaria,
China, Croatia, Czech Republic, Germany,
Hungary, Japan, Republic of Korea, Latvia,
Macedonia, Poland, Serbia and Montenegro,
Slovak Republic, Slovenia, Switzerland,
Taiwan.
Denmark, Finland, Norway, Sweden.
Armenia, Azerbaijan, Belarus, Georgia,
Kazakhstan, Kyrgyz Republic, Moldova,
Mongolia, Russia, Ukraine, Uzbekistan.
39 This classification is taken from Djankov, McLiesh & Shleifer, supra note 15, Appendix A. The
other studies mentioned in Part II.B use almost identical classifications.
2007] M.M. SIEMS RECONCILING LAW & FINANCE AND COMPARATIVE LAW
65
Yet these and similar classifications are not at all self-evident. Rather, for about
eighty per cent of the 129 countries that Djankov, McLiesh & Shleifer examined, the
categorization according to legal origin is far from clear. The difficulty arises mainly
with respect to legal systems in Eastern Europe, Asia, Africa, and Latin America. The
following examples illustrate this difficulty.
According to Djankov, McLiesh & Shleifer, Lithuania belongs to the French legal
family and Latvia to the German legal family because Latvias laws belonged to the
German civil law tradition prior to annexation by the Soviet Union in 1940; it
reverted back to them in 1991. Lithuania was influenced by French and Dutch law
both before its annexation in 1940 and after independence in 1990.40 This is,
however, hardly obvious. Although the Latvian and Lithuanian civil codes have been
influenced by German and French law, they are not at all mere copies of the parent
codes. Rather, they are drafted in a comparative fashion that takes different models
into account. For instance, though the Latvian Civil Code41 bears some resemblance
to the German Civil Code, it is unclear in what proportion it is based on the German
model. More generally, legal traditions beyond the German and French influenced
Latvian and Lithuanian law. In recent years, both legal systems have been influenced
by legal advice from the Nordic countries and the United States, as well as by the
implementation of European directives.
More importantly, the notion of legal families is not onlyor perhaps not even
primarilyabout legal rules as such. If one looks at other criteria it becomes clear
that Latvia and Lithuania should not necessarily be placed in separate categories. The
Djankov, McLiesh & Shleifer categorization disregards historical similarities between
the Baltic States such as the occupation by Germany and the Soviet Union during
World War II, contemporaneous periods of independence after World War I and 1990,
and the accession to the European Union in 2004. This common history is reflected in
legal rules, culture, and practice. My own experience teaching students from the three
Baltic States suggests to me that their law and legal practice cannot be categorized
along lines established by German or French legal influence. Rather, distinctions such
as EU/non-EU, Eastern / Western Europe, and Baltic/non-Baltic countries emerged.
As a result, it appears strange that according to Djankov, McLiesh & Shleifer, for
example, Latvia and Taiwan are put into one legal category and Lithuania and Syria
into another. This insight can be extended to other Eastern European countries, whose
legal systems often have a mixture of different influences and exhibit features
dissimilar to those of Western Europe.
According to Djankov, McLeish & Shleifer, China and Japan are treated as being
of German legal origin. At least with respect to China, this does not make sense at all.
40 Ibid. at 10-11.
41 For an English version, see The Civil Law of Latvia, online:
[Vol. 52
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
66
While Djankov et al. give no reasons for their categorization, it likely results from
having traced some export of German law to China. For example, The Company Law
of the Peoples Republic of China42 of 1993 was primarily based upon the company
laws of Taiwan, France, Germany, and Japan. For language reasons, legislators paid
particularly close attention to the Taiwanese law. Yet, Taiwans company law is itself
a hybrid, since it was originally based on both German and Japanese law and, after
World War II, came under U.S. influence.43 As a result, codified Chinese company
law is to a large extent a mixture of various legal influences and not simply of
German legal origin. This can also be seen in other areas of Chinese law because, in
contrast to Germany (or France), there is no comprehensive civil code,44 and Chinese
securities law is in principle based on the U.S. model.45
It is more difficult to criticize the classification of Japan as being of German legal
origin. Between 1890 and 1900 Japan did indeed copy large parts of the five major
German codes.46 However, these legal transplants have not necessarily retained their
importance to Japanese law. For example, the Commercial Code of Japan has been
substantially changed since World War II, in particular because of American
influence.47 The same is true for other areas of trade and business law.48
42 Adopted at the 5th Sess. of the Standing Comm. 8th Natl Peoples Cong., 29 December 1993,
promulgated as Order No. 16 of the President of the P.R.C., 29 December 1993, effective 1 July 1994,
trans. in The Company Law of the Peoples Republic of China (Beijing: Foreign Language Press,
2001) (P.R.C.).
43 See Mathias M. Siems, Convergence in Shareholder Law (Cambridge: Cambridge University
Press) [forthcoming in 2007].
44 There are only two legislative sources. See General Principles of the Civil Law of the Peoples
Republic of China (adopted at the 4th Sess. of the 6th Natl Peoples Cong., 12 April 1986,
promulgated as Order No. 37 of the President of the P.R.C., 12 April 1986, effective 1 January 1987)
(P.R.C.), trans. by Chinacourt, online: Chinacourt
Peoples Cong., 15 March 1999, effective 1 October 1999) (P.R.C.), trans. online: Judicial Protection
of IPR in China
45 See Lawrence S. Liu, Chinese Characteristics Compared: A Legal and Policy Perspective of
Corporate Finance and Governance in Taiwan and China (2001) at 2, online: Social Sciences
Research Network
46 See e.g. Zweigert & Ktz, supra note 16 at 298-301. But see Masao Ishimoto, Linfluence du
Code civil franais sur le droit civil japonais [1954] R.I.D.C. 744.
47 See generally Curtis Milhaupt, Creative Norm Destruction: The Evolution of Nonlegal Rules in
Japanese Corporate Governance (2001) 149 U. Pa. L. Rev. 2083. Milhaupt writes:
[T]he validity of the classification scheme used [by certain authors] to create the legal
origin variable is highly suspect. For example, these studies list Japan as belonging to
the German civil law family. This is partially, but only partially, true of Japans five
major codes … But many subsequent Commercial Code revisions and [a number of]
important economic regulatory statutes [bearing on investor protections] are of U.S.
origin. German law has had only a minor influence on postwar Japanese legal
developments. Thus, the classification for Japan is only about partially accurate and no
theory is offered to explain why legal origin, as opposed to subsequent legal
67
2007] M.M. SIEMS RECONCILING LAW & FINANCE AND COMPARATIVE LAW
Yet the more fundamental counter-argument is that it is not enough to look only
at legal rules. In the light of the deeper structure of legal systems, the distinction
between the common law and French, German, and Nordic civil law begins to appear
both overly legalistic and Eurocentric.49 In particular, one must consider legal culture
if one hopes to obtain a meaningful understanding of different legal traditions. Legal
culture refers to those elements in law that go beyond the mere content of statutory or
case law. It includes the historical background of a legal system, the emergence of
sources of law, the systematization of the law, the style of argument and codification,
legal education, and the ranking of law in a countrys social order.50 The importance
of these factors should not be underestimated. It is sometimes even claimed that
because of these factors, legal transplants are impossible in the sense that even
formally identical rules, being interpreted and applied differently in different legal
systems, do not survive the journey from one legal system to another unchanged.51 In
any case, it is necessary to take into account the characteristics of Asian legal
traditions52 in order to avoid the misleading results that flow from superimposing
European legal traditions onto non-European countries.
The need for a more tentative approach is confirmed in the case of Africa. Based
on the history of colonization, Djankov, McLiesh & Shleifer regard, for example, the
former English colonies of Botswana and Ghana as countries of English legal origin
and the former French colonies Mali and Niger as countries of French legal origin.
There is, however, no reference to the former German colonies, such as Namibia and
Togo. The reason for this is probably that the German colonial regime did not last
very long (18841919)53 and that, therefore, law and legal culture in Namibia and
Togo may not have changed significantly and permanently. Yet, one can also question
the impact of other European legal traditions in other African countries. A simple
categorization according to English common law or French civil law once again
developments, would be determinative of corporate governance patterns. It would not
be surprising if the classifications of legal origin for other countries in the study were
subject to similar defects (ibid. at 2123, n. 131).
48 See R. Daniel Kelemen & Eric C. Sibbitt, The Americanization of Japanese Law (2002) 23 U.
Pa. J. Intl Econ. L. 269.
49 See Boaventura de Sousa Santos, Towards a New Common Sense: Law, Science and Politics in
the Paradigmatic Transition (New York: Routledge, 1995) at 273; Mathias Reimann, The Progress
and Failure of Comparative Law in the Second Half of the Twentieth Century (2002) 50 Am. J.
Comp. L. 671 at 685; Upendra Baxi, The Colonialist Heritage in Pierre Legrand & Roderick
Munday, eds., Comparative Legal Studies: Traditions and Transitions (Cambridge: Cambridge
University Press, 2003) 46 at 49; Gnther Frankenberg, Critical Comparisons, Re-thinking
Comparative Law (1985) 26 Harv. Intl L.J. 411 at 422, 442.
50 See e.g. David Nelken, Disclosing/Invoking Legal Culture: An Introduction (1995) 4 Soc. &
Leg. Stud. 435 at 438.
51 See Pierre Legrand, The Impossibility of Legal Transplants (1997) 4 M.J.E.C.L. 111.
52 See e.g. H. Patrick Glenn, Legal Traditions of the World, 2d ed. (Oxford: Oxford University Press,
2004) at 301-42 [Glenn, Legal Traditions].
53 See generally William Otto Henderson, The German Colonial Empire, 18841919 (London:
Frank Cass, 1993).
[Vol. 52
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
68
disregards deeper legal structures, such as the question of how courts work or how
new law, old law, and customs interact. Thus, the fundamental question of how the
imposed new legal traditions mixed with chthonic and Islamic legal traditions54 has to
be answered in order to get to a meaningful description of legal families in Africa.55
This is not merely of conceptual importance. For instance, in Islamic commercial law
there are Islamic partnerships but no corporations because the concept of separate
legal personality is unknown.56 Thus, for certain issues, such as shareholder
protection in different countries,57 there is a fundamentally different starting point in
comparison with the commercial law of the West. The law and finance categorization
is therefore too heavily focused on European or Western legal traditions. A more
significant distinction would be between the concept of law in the West, which may
be characterized by its Christian roots, specific form of rationality, and concept of
rights, and the African legal traditions that continue to exist despite colonization.58
Latin America may be less problematic. Law and finance studies treat the Latin
American countries as being of French legal origin. It is indeed correct that via Spain
and Portugal almost all Latin American countries were influenced by the civil law
tradition as exemplified by the French Civil Code. Yet even law and finance scholars
see differences between the European parent tradition and its Latin American
offspring. In particularciting Merryman59it is said that the export of the
Napoleonic Code had pernicious effects in French, Belgian, Dutch, Spanish, and
Portuguese colonies that it did not have in France itself.60 Consequently, the fact that
the French legal family performs worst61 in most law and finance studies is mainly a
reflection of Latin American performance. Yet, it is doubtful whether this
performance is really linked with the export of French law to these countries. It is
54 See Glenn, Legal Traditions, supra note 52 at 59-91, 170-221.
55 Similarly, in reference to the spread of the common law, Zweigert & Ktz write:
[T]his might lead one to the conclusion that in the areas of Africa which were
previously under British rule most legal relations today are governed by the rules of
English Common Law. This conclusion would be wholly erroneous. The fact is that to
much [of] the largest part of the African population the Common Law is of almost no
practical significance; the legal relations of Africans, in contract and land matters as
well as family and succession matters, are principally governed by the rules of
customary African law, and in many regions also by the rules of Islamic law (supra
note 16 at 230).
56 See Glenn, Legal Traditions, supra note 52 at 183-84.
57 See e.g. La Porta et al., Law and Finance, supra note 13.
58 Cf. Glenn, Legal Traditions, supra note 52 (So we should … start thinking about the common
law and the civil law as representing some of the same ideas, compared with other traditions at 166).
59 John Henry Merryman, The French Deviation (1996) 44 Am. J. Comp. L. 109 at 116.
60 See Beck & Levine, supra note 10 at 259.
61 See Part I.B, above. For a counter-reaction to this result by the French Ministry of Justice, see
Attractivit conomique du Droit, Programme international de recherche, online: Mission de
Recherche Droit et Justice
des Amis de la Culture Juridique Franaise, supra note 6.
69
2007] M.M. SIEMS RECONCILING LAW & FINANCE AND COMPARATIVE LAW
likely that legal and economic problems would not be significantly different had
Latin American legal systems not copied the French but, for instance, the German
Civil Code. This can also be seen by the fact that German influence on Brazilian law62
and U.S. influence on the commercial law of most Latin American countries63 have
not caused an automatic change for the better. Thus, the problems in Latin America
seem to have a greater connection with the social and economic impacts of
colonization64 than with the borrowing of particular foreign statutes.65
As a result, one cannot escape the conclusion that the law and finance
categorization of most countries of the world according to a small number of legal
families is to a large extent arbitrary. This is also in line with insights from general
comparative law, as even comparative lawyers who still apply the notion of legal
traditions66 emphasize the notions limits. For instance, legal traditions are said to be
just a loose conglomeration of data,67 the idea of legal families is used purely for
explanatory purposes,68 and even if [w]e mostly continue to divide the world into
civil law, common law, and several other systems[,] … we know that these are ideal
types which merely serve our need to maintain a rough overview.69 Thus, the
criticism in this part does not imply that one should altogether cease to talk about
comparative law in terms of legal traditions or families. It makes sense for a textbook
on comparative law to divide its chapters according to such categories, as both
62 See Zweigert & Ktz, supra note 16 at 115.
63 See M.C. Mirow, The Code Napolon: Buried but Ruling in Latin America (2005) 33 Denv. J.
Intl L. & Poly 179 at 185-87.
64 Glenn suggests a possible reason:
[I]n creating large states, large corporate structures, large labour organizations, large
legal professionsin short, large institutionalized lites in all directions, western law
provides all the disadvantages of a large, wooden house in a warm, humid climate. It
may be beautiful, and well-designed, but be subject to many forms of internal rot. To
survive, it requires protection beyond the structure itself and if this is neglected, or
impossible, the structure will not last (Legal Traditions, supra note 52 at 265).
65 For a related point, see ibid. (Western development work has thus far been unable to overcome
the problem of widespread corruption of western institutions and western law when it has been
transplanted abroad … There appears to be little difference between civil and common law traditions in
this regard at 267-68).
66 For examples of comparative lawyers critical of the notion, see supra note 6.
67 Glenn, Legal Traditions, supra note 52 at 15.
68 Ren David, Les grands systmes de droit contemporains, 9th ed. (Paris: Dalloz, 2000) at 15,
trans. by Zweigert & Ktz, supra note 16 at 73. See also David & Brierley, supra note 16 (the idea of
legal families is no more than a didactic device at 21).
69 Reimann, supra note 49 at 677. Zweigert & Ktz express a similar sentiment: [A]ny division of
the legal world into families or groups is a rough and ready device. It can be useful for the novice, by
putting the confusing variety of legal systems into some kind of loose order, but the experienced
comparatist will have developed a nose for the distinctive style of national legal systems … (supra
note 16 at 72).
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
70
Zweigert & Ktz70 and Glenn71 do. In these works, the authors can sufficiently
address the extent to which (1) the spread of a particular legal tradition has really
changed pre-existing traditions, (2) the imported tradition has been displaced by more
recent legal traditions, and (3) the secular and religiously inspired legal traditions
have interacted in the same country.72 However, for an econometric study such as
those performed by law and finance scholars, clear criteria are necessary because
otherwise measurement errors and biased coefficients can result.
[Vol. 52
III. Unbundling Legal Families: The Search for Characteristic
Features
The criticism put forth in Part II of this article does not imply that different legal
traditions are irrelevant. The general statement that a countrys legal heritage shapes
its approach to property rights, private contracting, investor protection, and hence
financial development73 makes sense. Yet, at least for an econometric study on law
and finance, one has to find more precise criteria than worldwide distinctions between
different legal families. This part identifies and discusses characteristic features that
are related but not identical to the notion of legal families.74
A. Unreliable Categories
First of all, given that Roman law is a foundation for civil law, but not for the
common law,75 one might expect that the relevance of Roman law to a particular legal
system is an appropriate criterion for analysis. Yet there are problems with this
proposition. On the one hand, no contemporary legal system is entirely based on
Roman law. This is true for even traditional civil law countries such as France and
Germany. Both France and Germany can be understood as having mixed legal
systems because apart from Roman law, the droit coutumier of tribes from Northern
70 Ibid. The authors have chapters discussing the Romanistic legal family, the Germanic legal
family, the Anglo-American legal family, the Nordic legal family, the law in the Far East, and
religious legal systems.
71 Legal Traditions, supra note 52. Glenns chapter titles include A Chthonic Legal Tradition, A
Talmudic Legal Tradition, A Civil Law Tradition, An Islamic Legal Tradition, A Common Law
Tradition, A Hindu Legal Tradition, and An Asian Legal Tradition.
72 See Glenn, ibid. at 32-58, 343-47; Zweigert & Ktz, supra note 16 ([T]he division of the worlds
legal systems into families, especially the attribution of a system to a particular family, is susceptible
to alteration as a result of legislation or other events, and can therefore be only temporary at 66
[emphasis in original]); Ugo Mattei, Three Patterns of Law: Taxonomy and Change in the Worlds
Legal Systems (1997) 45 Am. J. Comp. L. 5 ([L]egal systems never are. They always become at 14
[emphasis in original]).
73 Beck & Levine, supra note 10 at 254.
74 This line of reasoning is different from the counter-argument that refers to factors unrelated to
legal traditions, such as geography, culture or politics. See notes 33-35 and accompanying text.
75 See Zweigert & Ktz, supra note 16 at 75, 100, 133-34; Alan Watson, The Making of the Civil
Law (Cambridge, Mass.: Harvard University Press, 1981) at 4.
71
2007] M.M. SIEMS RECONCILING LAW & FINANCE AND COMPARATIVE LAW
France and Germanic sources of law influenced their legal systems.76 On the other
hand, various traces of Roman law can be found in common law countries. For
instance, in Glenns description of English law, there are repeated references to
Roman law, starting with the influence of canon law and concluding with the role of
the European Union.77 Given these nuances, it would be necessary to set a threshold,
such as fifty per cent for example, to determine whether a legal system could be
called Roman for the purposes of categorization. This would raise the larger
problem of how to determine whether a country contains the required level of
Roman law, a very burdensome (if not impossible) task. As such, the criterion of
Roman law should not be adopted.
Similarly, it would be difficult to ascertain, for instance, the Frenchness or
Germanness of a countrys civil code. Although it is true that many countries have
been influenced by the French and German codes, no country has a code that is
identical to the French or German model. Even the codes of countries with close ties
to France and Germany may look quite different. For instance, there is probably no
provision that is exactly the same in any two of the German, Austrian, and Swiss civil
codes. Furthermore, the structure, language, and legal concepts are quite different in
each of these three. Even for these countries, it would therefore be a very time-
consuming process to look at the more than two thousand provisions of each code,
identify the similarities and differences between these provisions, and establish a
threshold above which these three countries would fall into the same category. For
non-European countries, the fact that codes are usually drafted in a comparative
fashion further compounds this matter. Thus, the closeness to a specific code
criterion is also not a very effective standard.
Similar problems arise for the criterion of common law influence in a particular
legal system. Legal rules, culture, and practice of countries that are said to belong to
the common law family can differ significantly. For instance, one can make the claim
that it is far from obvious today that the United States really belongs to the common
law family because [i]n many respects US law represents a deliberate rejection of
common law principle, with preference being given to more affirmative ideas clearly
derived from civil law.78 For countries in which precommon law traditions are still
relevant, a common law identity is even more doubtful. An example might be India
76 See e.g. Zweigert & Ktz, ibid. at 75, 139. See also rc, Family Trees, supra note 8 ([A]ll
European systems can be better understood as overlaps at 363).
77 Legal Traditions, supra note 52 at 222 (discussing canon law), 226 (on the similarities between
the Roman and English judiciaries), 231-32 (regarding the teaching of Roman law in the twelfth
century), 244 (on comparative law in England of the seventeenth century), 246 (discussing Sir Francis
Bacon as an example of an English lawyer trained on the continent), 254 (on the Roman origins of
trust law), 255 (concerning the influence of continental commercial law), 256 (on civilian lawyers
testifying in common law courts), 257 (on Pothier as formal authority in England), 257-58, n. 124
(discussing EU influence). See also H. Patrick Glenn, La civilisation de la common law [1993]
R.I.D.C. 559.
78 Glenn, Legal Traditions, supra note 52 at 248.
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
72
because, despite the relevance of the common laws older traditions, indigenous law
may be too deeply rooted to be simply disregarded.79 Similarly, it is said about
Thailand that it
[Vol. 52
has had in its modern texture a real mixture of sources such as English Law,
German Law, French law, Swiss Law, Japanese Law and American Law …
alongside historic sources in existence since 1283, such as rules from
indigenous culture and tradition, customary laws and Hindu jurisprudence, still
to be found in some modern enactments.80
One would therefore need to establish a specific criterionsuch as whether the most
important judgments of English case law are regarded as good lawin order to
determine whether a country belongs to the common law family. Once again, this
would be very difficult if not impossible in practice.
B. Four Criteria for Effective Differentiation
A more fruitful criterion is whether a country experienced European colonization
in the sixteenth through twentieth centuries. This is generally a precise criterion,
although one has to clarify how to deal with short-term colonial regimes and how to
categorize countries that were colonies of two different colonial powers. This
criterion overlaps to some extent with the conventional legal family definition,
because colonies of the same country are usually treated as members of that countrys
legal family. The advantage of this approach, however, is that it looks at the impact of
colonization on a particular country as a whole and not only at legal transplantations
that may have been only superficial or temporary. That said, analysis based on
colonial background may still reveal that the particular way a new law was applied
influenced the development of the country. For instance, Glenn has said that English
technique generally involved a more hands-off approach, whereas [t]he French saw
a more universal role for a more universal French law … 81 This difference in legal
approach may explain differences between former English and French colonies.
A second potentially effective criterion is language. At first blush, this may
appear strange: because no language is better than any other, it should not, as such,
influence financial development.82 Yet language is a key determinant of how well
ideas travel between different countries. This is all the more true if one views
translations as necessarily imperfect;83 the phenomenon of international languages
79 Cf. text accompanying notes 66-72; Glenn, ibid. at 296-97 (relating to chthonic, Hindu, Islamic,
and Asian legal traditions); Baxi, supra note 49 at 51.
80 rc, Family Trees, supra note 8 at 364.
81 Glenn, Legal Traditions, supra note 52 at 259. Differences in colonial strategies are also
addressed in Daniel Berkowitz, Katharina Pistor & Jean-Franois Richard, Economic Development,
Legality, and the Transplant Effect (2003) 47 European Economic Review 165.
82 But see Association Henri Capitant des Amis de la Culture Juridique Franaise, supra note 6
(Gnie de la langue franaise? at para. 79).
83 See e.g. Glenn, Legal Traditions, supra note 52 at 47-48.
73
2007] M.M. SIEMS RECONCILING LAW & FINANCE AND COMPARATIVE LAW
(first Latin, then French, and now English) and the possibility of translations do not
ensure accurate communication between native speakers of different languages.
Though it can be overstated, there is no denying that commonality of language is
relevant. A common language facilitates not only the copying of black letter law and
its application in practice, but also the exchange of information about its
philosophical, sociological, and economic background.
Third, the relative importance of statutory law and courts serves as a useful
criterion. The traditional distinction between civil law and common law holds that
statutory law is more important in civil law countries and that courts are more
important in common law countries.84 Yet, as statutory law and courts coexist in all
countries today,85 determining which legal institution is more important becomes
difficult. A possible (though rather formal) standard would be the existence of a
comprehensive civil code that covers at least contract, tort, unjust enrichment, family,
and succession law. Such a model could be seen as an illustration of what lawyers in a
particular country expect a proper legal system to look likenamely, an
extensively codified system logically structured by abstract legal norms. A more
material criterion would be, for instance, the independence and power of judges and
the correspondingly reduced influence of statutory law. In this respect, law and
finance research on the political channel can be used in classifying different
countries.86
Fourth, a similar but not identical factor for consideration is the formality or
flexibility of a legal system. Although it can be suggested that formality is linked to
statutory law and flexibility to judge-made law, this is not necessarily the case, since
statutory law can also be flexible and case law rigid. The relationship between this
consideration and that of legal tradition is particularly strong in the case of French
civil law, which is regarded as more formal than other civil law systems.87 As a proxy
for formality or flexibility, the law and finance research on the adaptability channel
offers useful insight for the classification of different countries.88
IV. Re-examining the Differences
My suggestion is to use the four identifiers considered in Part III.B instead of the
legal origins classification used in the law and finance studies. I will therefore provide
an example of the findings of a study conducted using the legal origins classification
and contrast it with my own approach.
84 See e.g. Zweigert & Ktz, supra note 16 at 69, 71.
85 See e.g. Roe, Legal Origins, supra note 22 at 475-79.
86 See generally note 28 and accompanying text.
87 See e.g. Beck & Levine, supra note 10 at 259. See also Roe, Legal Origins, supra note 22 at
489-92 (noting that over-regulation in civil law countries differs between labour and securities law).
88 See generally note 30 and accompanying text.
74
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 52
A. An Example from the Legal Origins Literature
The Djankov, McLiesh & Shleifer study on creditor protection identified four
variables as proxies for the strength of creditor rights.89 The laws of 129 countries
were coded according to these four variables, and each country was given a score
from 0 (signifying weak creditor protection) to 4 (signifying strong creditor
protection). On the basis of these ratings, Djankov, McLiesh & Shleifer calculated the
mean of creditor rights for each legal origin. The result is shown in Table 1, which
demonstrates the typical90 result that the average score of countries of English legal
origin (2.222) is higher than the average score of countries of French legal origin
(1.328). Because it is possible that differences between the two groups exist simply
by chance, the authors carried out the so-called t-test to ascertain that the difference
in means is statistically significant.91 In this case, the t-value was calculated to be
3.721; the difference is significant at the one per cent level, leading Djankov,
McLiesh & Shleifer to infer that countries of English legal origin really do provide
better creditor rights than those of French legal origin.
89 These four variables are: (1) restrictions on a debtors right to file for bankruptcy or restructuring,
(2) secured creditors ability to seize collateral after the bankruptcy or restructuring petition is
approved, (3) secured creditors right to be paid first out of the proceeds of liquidating a bankrupt firm,
and (4) management removal from administration of its property pending the resolution of the
restructuring (Djankov, McLiesh & Shleifer, supra note 15 at 25).
90 See Part I.B, above.
91 The t-value yielded by the calculation correlates with the significance of the difference of the
means. A t-value below 1.28 means that the difference of means is not statistically significant. A t-
value between 1.28 and 2.58 indicates that the difference is statistically significant at the ten per cent
level (i.e., that there is a ten per cent probability that the means are different simply by chance). A t-
value greater than 2.58 indicates significance at the one per cent level. The higher the t-value, the
greater the likelihood that the means are different not simply by chance. Tables illustrating the
relationship between t-value and significance can be found in any econometric textbook. See e.g.
James H. Stock & Mark W. Watson, Introduction to Econometrics (Boston: Addison Wesley, 2003) at
644.
2007] M.M. SIEMS RECONCILING LAW & FINANCE AND COMPARATIVE LAW
75
Table 1: Creditor Rights by Legal Origin (Djankov, McLiesh & Shleifer)92
Legal Origin
Number of
countries
Mean of
creditor rights
English
French
German
Nordic
Socialist
All
36
63
16
4
10
129
2.222
1.328
2.333
1.750
2.182
1.789
t-test
English vs.
French:
3.721
(i.e.,
significance
at the 1 %
level)
B. An Alternative Methodology
The comparison of means according to legal origin is not particularly illustrative,
because the way law and finance scholars assign countries to different legal origins is
to a large extent random.93 Instead, I have used the Djankov et al. data on creditor
rights and grouped the 129 countries according to colonizing power, language,
relative importance of courts and statutory law, and formality and flexibility of
legal systems as proxies for legal origin. This is shown in Table 2 and subsequently
explained.
92 Supra note 15 at 26 (Table II.A for 2003).
93 See Part II, above.
76
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 52
Table 2: Different Criteria
(1) Creditor Rights by Colonizing Power
Countries
Number of
countries
Mean of
creditor
rights
Former English colonies
Former French colonies
Former Spanish colonies
Former Portuguese colonies
Others
28
24
18
3
59
2.179
0.75
1.667
1.33
2.067
(2) Creditor Rights by Language
Countries where people speak
English
French
Spanish
Portuguese
Arabic
Scandinavian languages
German
Others
Number of
countries
Mean of
creditor
rights
27
16
19
4
14
4
3
51
2.380
0.5
1.684
1.25
1.5
1.75
2.333
2.059
t-test
Former
English vs.
former
French
colonies:
4.3853
(i.e.,
significance
at the 1 %
level)
t-test
English vs.
French:
5.384
(i.e.,
significance
at the 1 %
level)
2007] M.M. SIEMS RECONCILING LAW & FINANCE AND COMPARATIVE LAW
77
(3) Creditor Rights by Supreme Court Power94
Countries with
Number of
countries
Control of administrative cases
No control of administrative
cases
Term of supreme court judges of
at least 6 years
Term of 2 to 6 years
Term of less than 2 years
46
14
49
9
2
(4) Creditor Rights by Legal Justification95
Mean of
creditor
rights
2.086
1.714
2
2.111
1.5
Countries with
Number of
countries
Mean of
creditor
rights
Flexible legal justification (0)
Partly flexible legal justification
(0.33)
Partly inflexible legal
justification (0.66)
Inflexible legal justification (1)
6
14
28
23
2.5
2.5
1.821
1.565
t-test
control vs.
no control:
1.034
(i.e., no
significance)
t-test
partly
flexible vs.
partly
inflexible:
1.827
(i.e.,
significance
at the 10 %
level)
94 The criterion of supreme court power is based on La Porta et al., Judicial Checks and Balances,
supra note 29 at 34-35.
95 The legal justification criterion is based on Djankov et al., supra note 31. It is formed by the
normalized sum of (i) complaint must be legally justified, (ii) judgment must be legally justified, and
(iii) judgment must be on law (not on equity). The index ranges from zero to one, where higher values
mean a higher use of legal language or justification (ibid. at 465).
78
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 52
1. Methodology
Table 2 uses the same methodology as Table 1.96 As such, Table 2 not only
categorizes countries into different groups but also examines whether the differences
in creditor protection affiliated with these groups are statistically significant. The
result is that the first two categories (colonizing power and language) have a high
statistical significance at one per cent, which means that we can be ninety-nine per
cent certain that the differences are not simply random. By contrast, the third category
(supreme court power) is not statistically significant, and the fourth category (legal
justification) is only significant at the ten per cent level. While this does not mean that
these latter two categorizations are useless, we must at the very least interpret the
differences in these categories with caution.
2. Explaining the Categories
The categorization of countries according to colonizing power and language(1)
and (2) of Table 2resulted in some countries being classified as others. This was
necessary because some countries either have never been colonies or do not belong to
one of the seven language groups listed as part of this table. Moreover, some
countries (e.g., Togo, Burundi, Rwanda, and Tanzania) have been colonized by
various powers so that they could not be categorized in this respect. With respect to
language, a slightly different approach was taken. As the use of the same language
fosters communication about law and legal practice, I considered it sufficient if one of
the official languages was English, French, Spanish, or Portuguese. Only when a
double assignment was necessary, and neither of the two languages was dominant,
was a country classified as one of the others.
To categorize countries according to the relative importance of statutory law and
courts and to the formality and flexibility of legal systems, the proxies supreme court
power and legal justification(3) and (4) of Table 2were used. The data derive
from La Porta et al. and Djankov et al. and have not been re-examined. The role of
the supreme courtin particular its powers and the independence of its judgesis a
good proxy for the attitude with which a particular legal system regards courts in
general. Similarly, the way courts justify a decision is a suitable proxy for the
formality or flexibility of a legal system in general. Of course, the use of this data
does not imply that, for instance, the flexibility of the law of a particular country is
perfectly measured by this single variable. However, given the fact that more than
sixty countries are taken into account, these factors can provide some indication of
the differences between legal systems.
96 See Part IV.A, above.
2007] M.M. SIEMS RECONCILING LAW & FINANCE AND COMPARATIVE LAW
3. The Results
79
The general result shown in Table 2 is that there are considerable differences
between groups of countries. This suggests that legal rules are not merely
autonomous tools of governance but are significantly influenced by certain elements
of a countrys particular legal tradition.
Of the four criteria, colonizing power and language(1) and (2) of Table 1
appear to be more important than supreme court power and legal justification(3)
and (4) of Table 2for determining creditor-rights protection. Yet, the high statistical
significance of the first two criteria may also be explained by the fact that the
questions that determine the creditor-rights score97 were drafted by English-speaking
lawyers from one of the former English colonies (namely, the United States).98 The
analysis could therefore simply confirm the fact that ideas spread more easily among
countries with the same language. Thus, the third and fourth criteria are also
important because they may reflect legal culture at a deeper level.
In order to understand the results it is also useful to look at the places where these
four identifiers overlap. The classifications based on colonizing power and language
are similar but not identical. As the former looks at history and the latter at the present
situation, some subgroups of the first identifier have fewer members. In this respect,
language is more precise because, for instance, it takes into account the fact that ties
with former occupiers may have weakened in Arabic-speaking countries. That said,
the criterion of colonizing power can have the advantage that the former colonial
powers are not themselves included, and thus the category may pinpoint problems
that are specific to former colonies.
With respect to supreme court power, all of the former English colonies (and
English-speaking countries) belong to the forty-six countries where there is supreme
court control of administrative cases and to the forty-nine countries where the term of
supreme court judges is more than six years. Former English colonies aside, there is a
mixed picture for the supreme court control of administrative cases. The fourteen no-
control countries are former French, Spanish, and Portuguese colonies as well as
countries where people speak French, Portuguese, Spanish, Arabic, German, and the
Scandinavian languages. Regarding the term of supreme court judges, more than six
of the eleven countries where judges are appointed for less than six years are former
Spanish colonies or are Spanish-speaking countries.
The results for legal justification are even more diverse. As expected, the English-
speaking countries (as well as the former English colonies) are flexible in this respect
(mean: 0.42).99 Yet, the score for the Nordic countries implies even slightly more
97 See Djankov, McLiesh & Shleifer, supra note 15 at 25.
98 This home bias is also a problem of other law and finance studies. See Lele & Siems, supra note
19; Siems, A Critique on La Porta et al., supra note 32.
99 The following numbers in parentheses all refer to means. A mean closer to 1 indicates less
flexibility.
[Vol. 52
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
80
flexibility (0.418). The results for French- (0.668) and Arabic-speaking (0.736)
countries as well as for the former French colonies (0.716) indicate that these
countries are somewhat inflexible. Finally, German- (0.89), Spanish- (0.903), and
Portuguese-speaking (1) countries as well as the former Spanish (0.897) and
Portuguese (1) colonies are least flexible.
This article cannot explain in detail why there are differences in these four
identifiers. Yet, it is apparent that these differences confirm the danger of using one
narrow criterion, such as the common law / civil law distinction. The results by
Djankov et al. and other law and finance scholars who use this distinction (Table 1)
are most similar to my results when the proxies colonization and language are
used (Table 2). My approach has, however, the advantage of clearly identifying the
basis for categorization. If one speaks of colonies (and not members of the same legal
family) it becomes clear that the reasons for differences among countries may be a
consequence not only of different legal origins but also, for instance, of the ongoing
political and economic effects of colonization.100 The same is true for language
because language does not only foster communication about the law but is also a
transmitter for ideas in general (memes).101
Law and finance scholars also misinterpret how legal thinking affects the law of
different countries. They take as a given that because countries belong to a particular
legal family one has to examine the channel through which these different legal
traditions shape finance.102 Yet legal traditions are not unvariegated wholes; they have
differing components. One of them is, for instance, whether legal systems follow a
formal or flexible way of legal reasoning. The adaptability of legal systems is
therefore important not as an adaptability channel, but rather as a criterion that can
help identify commonalities and differences in legal thinking among countries. Such
a criterion obviates the need to rely on preconceived categories of civil and common
law.
Conclusion
[C]omparative [law] is bound to be superficial.103 Therefore, to argue that law
and finance studies employ categories that are not one hundred per cent accurate is to
mount an unfair criticism. There could also be good reasons why the reduction of
100 See Part II, above.
101 On memes, see especially Richard Dawkins, The Selfish Gene, 2d ed. (Oxford: Oxford
University Press, 1989). The inventor of the concept of memes was probably Richard Semon. See
Richard Semon, Die Mnemische Empfindungen in ihren Beziehungen zu den Originalenempfindungen
(Leipzig: Wilhelm Engelmann, 1909). See also Simon Deakin, Evolution For Our Time: A Theory of
Legal Memetics (2002) 55 Current Legal Probs. 1.
102 See text accompanying notes 27-30.
103 Lawson, supra note 1 at 16.
81
2007] M.M. SIEMS RECONCILING LAW & FINANCE AND COMPARATIVE LAW
complexity resulting from this kind of numerical comparative law104 could be a way
forward. However, with respect to the specific question of legal traditions, these
studies are questionable. The worldwide distinction between different legal families,
such as common law and (French, German, or Nordic) civil law is not useful for law
and finance analysis. Instead, more precise criteria must be found. My proposal is that
the categories colonizing power, language, relative importance of courts and
statutory law, and formality and flexibility of legal systems should be used. This
article has provided an example of what this new approach could look like. Future
research may also apply these criteria for statistical regressions.105 Such studies would
enable researchers to analyze with greater nuance the determinants (legal and
otherwise) of important phenomena like investor protection or World Cup success.
104 See generally Mathias M. Siems, Numerical Comparative Law: Do We Need Statistical
Evidence in Law in Order to Reduce Complexity? (2006) 13 Cardozo J. Intl & Comp. L. 521.
105 For the taking into account of specific econometric problems, such as endogeneity, see e.g. Stock
& Watson, supra note 91 at 333.
