Friday, November 28, 2014

Dialogus de Beijing Consensus -- Pessimo Response to Optimo on Experimentalism

Some responses to Optimo:

I think the problem of identifying ends is much more critical in the context of experiementalism than it is in the context of best practices (e.g., the Washington Consensus), because of the former's critical reliance on ex post evaluation.  If we can't agree on the ends, then we really can't do that evaluation, and experimentalism becomes reduces to simple decentralization.

Optimo appears to acknowledges the special difficulties that such normative questions pose for experimentalism, but suggests that that model can still survive.  In this, I question whether Optimo underestimates how domineering normative issues are in the context of law and development (as contrasted again, for example, developmental economics).  Law is a strongly normative phenomenon, and even the most hard-core positivists (like Pessimo) seem to have great difficultly separating the normative from the positive / procedural.  Almost all of the law and development projects that I am aware of have ultimately been informed strong normative understandings.  I would therefore at least hypothesize that experimentalism distinct difficulty with the normative may be a significantly more problematic feature in the context of law and development than it is in other areas of development. 

But then again, Optimo always has been more optimistic than Pessimo!

In responding to my skepticism regarding the suitability of experimentalism to national-level law and development projects, Optimo suggests "the beauty of experimentalism is the fact that we may arrive at different solutions through the same process, thus being able to share lessons while at the same time remaining context-dependent."  I assume this is related to Pessimo's subsequent discussion of meta-principles (attributing this, as Optimo likes to do, to that fictitious literary alter-ego of hers whom she calls 'Mariana Prado'.)  I'm not completely clear how these might work, but I do not see anything occurring in or coming out of China that fits this particular description -- certainly they have not been identified in the experimentalist literature referencing China.  Of course, this is ultimately an empirical question, and therefore beyond the terms of our dialogue (which are primarily conceptual). 

But then, Optimo does herself raise this empirical issue when she discusses the work of Chenggang Xu, who-- like many --identify Chinese experimentalism with numerous, spontaneous local rural land reform initiatives that took place in the 1970s.  As noted above, this is a commonly heard trope -- but there are a number of major problems with it.  First, this wasn't really experimentalism, it was reversion.  China had a regime of private agrarian land use rights in the 1950s, these so-called 'experiments' really simply reintroduced that regime.  It did not really develop anything new, which seems to be the raison d'etre of experimentalism as a modelRelatedly, one could also argue that these reforms were primarily an expression simply of gradualism rather than experimentalism.  There was never any question but that Deng was going to bring to China the 'free' market structure it had enjoyed in the early 1950s and found in most capitalist countries.  Again, this is not a question of experimentalism -- there was already much experience with these reforms, there really wasn't that much variation from locale to locale, and they were known commodities.  In fact, spiritually, these local experiments had much more in common with the Washington Consensus than they did with experimentalism.  Moreover, China's gradualism was ultimately dictated by political concerns rather than developmental concerns, and for this reason too would not seem to represent a developmental model. 

Sunday, November 16, 2014

Dialogus de Beijing Consensus -- The Beijing Consensus reloaded: Optimo on Experimentalism


We turn now to a second version of the argument that the so-called Beijing Consensus potentially represent a development model that contrasts with the Washington Consensus. This second version is based on the idea that this model is primarily based on experimentalism. 

Pessimo's previous post does not address whether there is experimentalism in China or somewhere else. Instead, Pessimo offers a challenge to the idea that experimentalism could potentially serve as a model for development. 

Pessimo's first argument against experimentalism as a model is the fact that it does not help us deal with the normative questions of development. In other words, experimentalism does not help us define the ends of development. Thus, its utility would be limited to situations in which the ends are already defined, which are often not the major problem that developing countries are confronting, according to Pessimo. 

Optimo's first reaction to Pessimo's point is to ask whether it would be even possible to develop a model to define ends. As Pessimo acknowledges, any attempt to define the goals of development and what a society is ultimately aiming at achieving, is highly dependent on the political, cultural and social context. Thus, abandoning the idea that there could or should be a model to define the ends of development may actually be the beginning of a much more promising conversation than the one that the development field has experienced thus far. Indeed, Optimo would argue that the fact that experimentalism seems to have abandoned the intention to define ends, and has focused instead on means, is not only a differentiating feature of this model over the Washington Consensus model, but it may actually be considered a significant upgrade, as development models go. It is a model that is aware of its limits and operates within realistic boundaries.

This is not to say that the search for a way to define and determine the ends of development is futile and should be abandoned. On the contrary, the entire development enterprise operates under the assumption that there is some search for a common goals of some sort, lest the concept of development becoming so diluted as to encompass "anybody's notion of utopia". But it may as well be the case that the model that will help us search for the ends is not and should not be the same as the one that will help us deal with the means. Thus, Optimo would advance the idea that experimentalism can still survive as a model, despite not helping us with the pressing normative questions that Pessimo correctly highlights in the last post. 

The second point raised by Optimo is the incompatibility between development in a national legal system and experimentalism: the former would intrinsically involve a top-down process that is antithetical to the central tenants of experimentalism. Optimo also disagrees with the premise of this statement, as it fails to distinguish between form and substance. One could argue that the substance of experimentalism is indeed context dependent, but the form is not. Thus, there is a dimension of the process that remains top-down, which is the formal dimensions -- i.e. what is the process through which actors may find the appropriate means to achieve predefined ends. The beauty of experimentalism is the fact that we may arrive at different solutions through the same process, thus being able to share lessons while at the same time remaining context-dependent. 

Regarding this last point, there is a recent article by Kevin Davis and Mariana Prado suggesting that experimentalism and other recent theories represent a move away from substantive commonalities, towards meta-principles. Since context-dependency prevents a productive conversation about shared lessons and transplanted solutions, the conversation may be more productive moving from the substance of development policies to its form. The idea of meta-principles captures this move. Acknowledging that, law and development theories are now focusing on the procedural and formal aspects of reforms, rather than the content. This is exactly what experimentalism seems to be doing. So, there is still a role for the "development facilitator located internationally", in the words of Pessimo.

Regarding the national legal system and its compatibility with experimentalism, Pessimo also argues that national policies would not be possible, because experimentalism is intrinsically connected with decentralization. Optimo thinks that the experimentation and decentralization may go together or not. And to make that point with an specific example from China, it is interesting to take a look at the work of Chenggang Xu from the University of Hong Kong. Xu argues that decentralization and experimentation are different features. While both have been present in China, their combination created a particular type of decentralization that is unique to that country (see the published version here and a previous draft that can be read for free here). 

What are the examples of experimentalism? One example was land reform in the late 1970s, which was initially done at the local level. According to Xu, when the local experimentation "was endorsed by the central government, they were implemented by all levels of government nationwide". Contrary to Pessimo, Optimo does not see any problem in calling this experimentalism. To be sure, Xu indicates that such experimentation was more prominent at the earlier stages of reform, and was significantly reduced as higher levels of economic growth were achieved. Still, the fact that such experimentation took place still suggest that there is something that could be potentially called the Beijing Consensus that would be based on experimentation.

In sum, Optimo remains and optimist that Beijing can offer something useful to other developing nations!

Thursday, November 6, 2014

Dialogus de Beijing Consensus -- The Beijing Consensus v.2.1: Pessimo on Experimentalism

A couple of years after Ramo postulated the notion of a Beijing Consensus, several scholars began advancing another kind of development model deriving, as they saw it, from China's experience.  One of these is Dani  Rodrik's 'New Development Economics', which -- inspired to considerable extent by Sebastian Heilmann's "Policy Experimentation in China's Economic Rise" -- "calls for an approach that is explicitly experimental, and which is carried out using the tools of diagnostics and evaluation. . . .The proof of the pudding is in the eating: if something works, it is worth doing." [p.2]  Does such experimentalism hold meaningful promise as a 'model' for promoting 'development' more generally?


'No, it does not.'

Following Optimo's advice regarding Ramo's model, we will leave alone the question as to how China's experience actually supports Rodrik's model.  

First off, I want to clarify what I see as the attributes of a 'model'.  A 'model' is by its very definition a universal template -- a developmental model is a template for development that can be effectively applied independently from context (unless some prerequisite context is specified in the model itself, which is not the case with 'new experimental economics'). Also, to the extent that 'law and development' models seem ultimately to be devoted to promoting development assistance, a developmental model also needs to provide direction for how such assistance should be constructed.  A developmental model that provides no significant guidance to the entity promoting development is not the kind of 'model' that serves the (traditional) 'law and development' agenda.

Following this, I would argue that Rodrik's model does not provide any meaningful guidance for developmental assistance, because it ultimately provide no universalizable template for developmental assistance -- at least in the area of law and development.

First, Rodrik’s experimental approach only really works when the developmental ends are agreed upon.  New Development Economics is a process by which we can identify different means that better serve some given ends.  But for many aspects of law and development, the means are the ends.  With regards to China, for example, the issues is often not ‘how best to promote rule of law’, but whether ‘rule of law’ should receive priority over other possibly competing goals – like ‘modernization’ or ‘economic growth’.  When the developmental issue is normative rather than practical, New Development Economics does not provide a model for promoting development.

Note that with regards to law, this problem can even infect aspects of legal development in which there appears to be consensus as to the developmental end.  For example, both the Chinese side and the development organization may agree that China needs to develop stronger ‘judicial independence’.  But for the most part judicial independence is as social construct -- it can mean very different things, both in terms of institutional structure and in terms of its contribution to the legal system, to different people.  One study suggests, for example, that donors are more likely to conceptualize judicial independence as means for promoting procedural justice, whereas Chinese are more likely to conceptualize judicial independence as a means of promoting substantive justice or factual accuracy in decisionmaking.  Again, experimentalism does not provide a model for addressing these kinds of developmental issues.

(One might respond that we could still use experimentalism to find ways of reaching and convincing the Chinese that judicial independence is indeed really about procedural justice rather than substantive justice.  But that gets us into question of moral imperialism, which Pessimo will leave for another time.)

I would suggest that a good many ‘law and development’ issues in China are of this sort.  There is in fact much disagreement between China – or at least China’s political leadership – and the law and developmental community as to what goals law and development should be striving to promote.  This is true not simply insofar as what following Randall Peerenboom we might call ‘thick’ notions of law and development is concerned, i.e., notions that see developmental as necessarily including promoting human and political rights, and various (liberal) conceptualizations of ‘justice’.  But it is also likely to be true with regards to more economically oriented developmental projects.  China and ‘the West’ often have significantly different understandings as to the role that market capitalism should play in society.  China’s understanding is much more akin to what is sometimes termed ‘economic nationalism’, while ‘the West’s’ understanding is much more politically neutral and cosmopolitan.  For this reason, even many economic legal reforms projects – such as reforms to corporate governance of capital markets – are ultimately thwarted by disagreement over ends, not by ignorance with regards to means.  Here, too, experimentalism does not help.

Another area in which experimentalism seems to have little utility is in that of national legal development.  As described by Rodrik, experimentalism is a very contextualized developmental process.  A national legal system, by context, is by its very nature a-contextual, it is almost by definition a ‘one-size-fits-all’ phenomenon insofar as the social space of the nation is concerned.  Here, a developmental model that focuses on responding to minute nuance of local context again is of little utility.  At most, experimentalism would counsel the developmental agent to promote decentralization, local experimentation, and perhaps data gathering.  But what would this accomplish?  Any local success or local failure could well hinge on contexts that are unique to the locale.  For this reason, local successes or failures are unlikely to be able to serve as positive or negative models for elsewhere, even after they have been ‘evaluated’ by central entities.  At best, decentralization and experimentalism might promote local legal development, but not national legal development.

This suggests that if there is a legal-developmental model in New Development Economics, it is a model that counsels the developmental agency to focus on local legal development rather than national legal development.  But there are problems with this as well.  One of the major value-added's provided by international legal developmental aid lies precisely in its greater familiarity with more global experiences with law and development.  The more localized and locally contextualized problems to be addressed through development, the less a developmental facilitator who is located internationally is able to bring to the table in the form of useable knowledge.  'Legal development' becomes reduced primarily to simply being a source of funds for local projects developed by local actors.  

Whether simply being a content-less source of funds this is sufficient to constitution a developmental model is an open question.  But whether this is a feasible model of development is another issue.  Such a model would be particularly costly to implement.  Development of local legal institutions costs about the same as development of national legal institutions.  But they benefit far, far fewer people.  Costs would be even further increased by the experimental nature of the development project, because it would involve funders paying for a lot of failures in addition to the occasional localized success.  Given its extremely high cost-to-benefits ratio, it seem very unlikely that many funders would embrace such a model.