sau's informal referee reports

Edward. Glaeser and Andrei Shleifer (2002), "Legal Origins," Quarterly Journal of Economics 117(4): 1193-1230.

There must be a million studies already published on legal origins. For yet another one to appear on a top economic journal, it'd better (1) offer a radically new theory, and (2) back it up with solid evidences. GS' theory is indeed radically new. They argued that the main difference between the civil and the common law systems was that in the former judges were state employees whereas in the latter judges were independent juries. All the other differences between the two systems allegedly came from there. What's the origin of such a difference? There's a tradeoff: independent juries were less biased (in favor of the king) but might be muzzled by local barons; state-employed judges were more biased but were also protected by the state from local harrassment. GS argued that during the 12th and 13th centuries, which they regarded as the crucial moment for the development of the two systems, England had a strong central government and was relatively peaceful, and hence independent juries was the efficient choice; France had a weak central government and was more chaotic, and hence judges were state-employed. (The "weaker" the central government was, the more it went out on a limb to protect judges against "strong" local barons... sort of make your head spin, right?)

Interesting theory. Did they back it up with solid evidences? First of all, the timing doesn't look right. Before the revolution, judges in France were probably as much captives of local barons as GS could imagine. "[J]udicial offices were regarded as property that one could buy, sell, and leave to one's heir on one's death. Montesquieu himself inherited such an office, held it for a decade, and sold it. The judges were an aristocratic group who supported the landed aristocracy [...] against the centralized governmental power in Paris. [...] [E]fforts by the Crown to unify the kingdom and to enforce relatively enlightened and progressive legislative reforms had frequently been frustrated. The courts refused to apply the new laws, interpreted them contrary to their intent, or hindered the attempts of officials to administer them. [...] [I]t was not unusual for continental judges to act much like their English counterparts. [...] [T]hey were interpreting creatively, building a common law that was a rival to the law of the central government in Paris and even developing their own doctrine of stare decisis." (Merrymen, The Civil Law Tradition, 1985) So much for state protection from local harrassment - and where was the alleged king-bias?

Second of all, codification of law, which GS regarded as a control device necessary only when judges were state-employed and hence happened only in civil-law countries, actually happened when the central governments were strong (and hence presumably the tradeoff should go in favor of independent juries) rather than when they were weak. This is true for both of the most important codifications in history: Corpus Juris Civilis and Code Napoleon. The history of the Roman Empire is especially illuminating. While GS would have predicted that its legal system would be closer to that of England, in fact "the adjudication of disputes fell more and more into the hands of public officials who were also learned in the law, but [whose] principal function was clearly understood to be that of applying the emperor's will. The judge had no inherent lawmaking power." (Merrymen, The Civil Law Tradition, 1985)

I don't want to be too harsh on GS. Legal systems are to certain extent deliberate designs by political elites, especially after a revolution. GS' theory might work better if they were talking about how the post-revolution system was in reaction to (the memory of) the pre-revolution horror. But, even if we try to save GS' theory in that way, the focus should be on why judges have a much smaller role in the civil law system, instead of on why judges are state-employed. (September 2004)


Edward Glaeser, Simon Johnson, and Andrei Shleifer (2001), "Coase vs. the Coasians," Quarterly Journal of Economics 116(3): 853-899. 

There are actually two seperate issues clumsily mixed together by these (apparently confused) authors. The first is whether it is desirable to regulate private contracts. The second is, provided some regulations are indeed desirable, whether a judge or a regulator will be a better enforcer of those regulations. Although the title of this paper misleadingly suggested that they're to focus on the first issue, what they studied was actually the second one. They regarded the main difference between a judge and a regulator as that the former was neutral whereas the latter was zealous. When enforcing regulations is costly, sometimes a zealous enforcer can do a better job.

They then used Poland and the Czech Republic to support their claim. Roughly speaking, Poland uses an SEC to regulate private financial contracts and thrives, whereas the Czech Republic uses judges and sucks. The latter suffers from poor corporate governance, where firm managers steal and tunnel assets at the expense of investors, which in turn results in its under-developed financial market.

Nice facts to know, but these facts are more relevant to issue #2. The paper never addresses issue #1: why regulations are desirable in the first place. To put it differently: why can't the Czech Republic firms and investors learn to sign more sensible contracts? The closest the paper got in answering this question is the following paragraph on p.893, which is evasive to say the least: "[More sensible] contracts bonding firms to treat investors well did not materialize. The survival of the theft-proof firms is not an efficient mechanism of economic selection. The most efficient firms might be the most attractive to tunnel, making them the least rather than the most likely to survive."

Anyone understands what the heck they were talking about please help me. (July 2004)