Dani Rodrik is well-known as a critic of ‘globalization’. Over a span of more than a decade, he has written numerous books, articles, and blog posts in which he takes issue with various aspects of international economic regulation. His latest effort is The Globalization Paradox: Democracy and the Future of the World Economy. For those of who have read his past works, much of the book will be familiar. In this new book, he attempts to synthesize many of the specific arguments he has developed over the years, and he also incorporates his thinking on the recent financial crisis. The result is several proposals for changes to the existing international architecture of globalization. In the final chapter of the book, he offers up what he refers to as ‘A Sane Globalization’, involving suggestions related to trade, finance, and labor migration. This review focuses on his discussion of the trade issue. I begin with a brief overview of Rodrik's description of the problems with the current trade regime and his suggestions for reform. I then offer a number of critiques of his arguments.

Rodrik starts the discussion of the problems he sees in today's trade regime by drawing a simple contrast between the GATT and the WTO. Put simply, in his view, the GATT was limited and good, whereas the WTO overreaches and is therefore bad. Looking at some specifics, Rodrik observes that under the GATT: (1) sensitive areas such as agriculture, textiles, and services were mostly excluded from the rules; (2) developing nations made few commitments to liberalize; and (3) import protection could be achieved through anti-dumping and safeguard measures. In addition to these substantive limitations to the GATT, Rodrik calls the GATT's enforcement powers ‘a joke’ and its obligations ‘patently unenforceable’ (pp. 72–74). As a result, any substantive obligations were of limited effectiveness.

While many people would (and did) consider these characteristics of the GATT to be flaws in need of correction, Rodrik sees them as positives. He attributes the GATT's ‘success’ to its ability ‘to achieve the maximum amount of trade compatible with different nations doing their own thing’, that is by keeping intrusion into domestic affairs to a minimum (p. 75).

By contrast, he argues, the WTO pursues ‘hyperglobalization’, with its rules now applying to the previously excluded sectors, and also to such areas as intellectual property, health and safety rules, and subsidies (p. 78). In this regard, he says that while GATT cases were ‘largely about tariffs and quotas’, WTO disputes ‘reach into domestic areas that were previously immune from external pressures’ (p. 79). He notes that under the GATT's ‘shallow integration’, the trade regime ‘require[d] relatively little of domestic policy’. By contrast, under the ‘deep integration’ of the WTO, ‘any discretionary use of domestic regulations can be construed as posing an impediment to … international trade’ and thus ‘[g]lobal rules in effect become the domestic rules’ (p. 83).

Rodrik later picks up this same theme when he talks about how globalization ‘gets in the way of national democracy’ (p. 190). He gives several examples in this regard, including the following: high labor, health and safety standards in some countries being undermined by low standards in other countries; and high tax rates in some countries being undermined by low rates in other countries (pp. 190–197).

He also mentions in this context one of his biggest concerns with the constraints imposed by international economic regulation: restrictions on the use of ‘industrial policy’ by developing countries as part of their efforts to achieve economic development. Industrial policy is a vague term, covering a lot of ground, so it is useful to look at the specific examples he gives: export subsidies, domestic content requirements, and the unencumbered use of others' intellectual property (pp. 198–199). In Rodrik's view, these are useful tools to promote industrialization, and the WTO's restrictions on such measures are therefore problematic.

Rodrik then presents what he calls the ‘Trilemma’ that results from the ‘tension between national democracy and global markets’, that is from the intrusion of the rules of globalization into the domestic policy-making arena. There are, he says, three options to deal with this tension. We can ‘restrict democracy’, by disregarding the wishes of the people of particular countries; we can ‘limit globalization’; or we can ‘globalize democracy’. But we cannot have hyperglobalization, democracy, and national self-determination all at once (p. 200). In this regard, Rodrik says that he is skeptical of ‘global governance’ because ‘there is simply too much diversity in the world for nations to be shoehorned into common rules, even if these rules are somehow the product of democratic processes’ (pp. 203–204).

He then turns to his proposals for remaking globalization, which involve limiting globalization so as to reduce the tension described. Chapter 11 sets out some key principles, and Chapter 12 contains specific proposals. The principles are as follows: 1. ‘Markets must be deeply embedded in systems of governance’; 2. ‘Democratic governance and political communities are organized largely within nation states, and are likely to remain so for the immediate future’; 3. ‘There is no “one way” to prosperity’; 4. ‘Countries have the right to protect their own social arrangements, regulations, and institutions’; 5. ‘Countries do not have the right to impose their institutions on others’; 6. ‘The purpose of international economic arrangements must be to lay down the traffic rules for managing the interface among national institutions’; and 7. ‘Non-democratic countries cannot count on the same rights and privileges in the international economic order as democracies’ (pp. 237–247). Of great importance here is principle 2, which holds that nation states are the foundation of governance. For Rodrik, globalization should not undermine this principle. Related to this is principle 4, which ensures that nations are able to regulate without intrusion from international rules and institutions. These two principles seem to guide his specific proposals.

Finally, he concludes with a chapter entitled ‘A Sane Globalization’, in which he sets out his vision for change to the trade regime. Here, he first notes that given what has already been accomplished in trade negotiations, there is not much liberalization left to do, so any gains would be minimal (p. 252). Instead, he says, future trade negotiations should be about ‘policy space’ (p. 253). A key element of this is an expanded ‘safeguard’ mechanism, which he refers to as ‘Developmental and Social Safeguards’. These safeguards would allow for trade restrictions and the suspension of WTO obligations – an ‘opt-out’ from the normal rules – for a number of purposes, such as: ‘distributional concerns, conflicts with domestic norms and social arrangements, prevention of the erosion of domestic regulations, or developmental priorities’. Trade restrictions would only be available where certain procedural requirements were met: the country ‘would need to demonstrate that it followed democratic procedures in reaching the determination that the safeguard measure is in the public interest’, with relevant criteria in this regard including ‘transparency, accountability, inclusiveness, and evidence-based deliberation’ (pp. 253–254). In this way, the safeguards would create ‘policy space’ for governments to regulate, and as a result would make the trade regime more stable and legitimate.

There are a lot of good things to say about Rodrik's writings on globalization issues. Unlike some critics of the current trade regime, he takes a very measured, thoughtful and balanced approach. Over the years, he has developed and refined his views, and in this book puts forward a set of policy proposals that might be taken up. He is not calling for a full-scale reversion to protectionism, as some critics of the system seem to want. Rather, he wants to see a moderated form of global trade regulation, which allows for more protectionism than he believes is currently permitted. Whether the substance of what he is proposing can be brought within the mainstream view of trade economists and policy-makers is unclear, but he is trying to make the best case possible for it.

With regard to the substance of his arguments, there are a number of general points that merit further discussion, some of which overlap. I address here the following: The role of policy space; the scope of the trade regime; and the relationship of democracy, sovereignty, and global governance.

Before turning to these issues, though, it is worth noting a couple more technical points. First, Rodrik's version of the history of the GATT and WTO may be somewhat misleading. In terms of the substance, the GATT was not as ‘shallow’ or ‘unenforceable’ as he indicates. On the issue of shallow versus deep integration, non-discrimination was one of the GATT's core principles from the outset, and Rodrik at times seems unaware of how broad the principle of non-discrimination can be. Such a requirement effectively reached all domestic policies, in the sense that all polices must not be discriminatory. That is, every measure a government takes in every area of regulation must comply with this requirement. Rodrik himself provides, later in the book, an example of the GATT's broad scope when discussing the Thailand–Cigarettes case (p. 195). But he does not seem to connect this case to his earlier discussion of the scope of the GATT.

As to the enforceability of the GATT, it has been argued that GATT rules were actually just as effective, in terms of usage and achieving compliance, as WTO rules (Posner and Yoo, Reference Posner and Yoo2005; for a response, see Davey, Reference Davey2006, and more generally Bown, Reference Bown2004 and Hudec, Reference Hudec1993). The instincts of most trade lawyers tell them that this is a bit of a stretch, and while empirical analysis of this issue is difficult, it is probably safe to say that WTO rules are, in important ways, more enforceable than were the GATT rules. Nonetheless, there is good evidence that GATT was not ‘a joke’, as Rodrik puts it, in terms of its enforcement mechanism.

Furthermore, the GATT was also not as successful as he makes it out to be. In fact, it was, in part, the substantive and procedural gaps he points out that led key GATT contracting parties to create the WTO. Presumably, these countries did so because they found the GATT rules to be lacking. If the WTO does pursue more intrusive globalization, it is because the governments wanted these new, expansive rules.

The converse of his mischaracterization of the GATT as extremely limited in nature is that the WTO is not as deep, enforceable, or unsuccessful as he makes it out to be. Without a doubt, integration under the WTO is deeper than under the GATT, but it is nothing close to the supra-national regulation and harmonization seen in the European Union. And while the enforcement mechanism is stronger than under the GATT, there have been a number of WTO disputes where enforcement procedures have not achieved results very quickly (or sometimes even at all, such as with the Hormones case he mentions a number of times). As to the overall success of the WTO, this is a difficult point to assess. However, despite troubles with the current negotiations, there are many aspects of the WTO, such as the dispute settlement mechanism, that have held up quite well.

Another technical point is that his argument about tariffs being quite low these days, and thus liberalization perhaps having only limited benefits, could be questioned. A quick look through World Tariff Profiles 2010 (WTO/ITC/UNCTAD, 2010) shows a wide range of tariffs in the double digits. Tariffs may not be as high as they used to be, but there are plenty of high ones. Thus, it is possible that additional tariff reductions could provide more significant economic benefits than he suggests.

Finally, Rodrik's focus is still on the WTO, and he pays little attention to the significant growth of bilateral and regional trade agreements. These agreements have gone well beyond WTO rules in many ways, including expanded intellectual property protection, investor-state dispute settlement on a variety of substantive investment issues, and enforceable labor and environment provisions. It would have been interesting to see these issues incorporated into his analysis. In addition, in the WTO context it may be possible to make the argument that rich countries control the agenda and force developing countries to accept unfavorable terms (although the role of developing countries has recently grown so as to raise skepticism about this argument). However, the spread of bilateral and regional trade agreements among a wide and varied group of developing countries undermines this point. The fact that developing countries are taking on obligations under these kinds of trade agreements suggests that they view themselves as benefiting from the commitments involved.

Putting these technical criticisms aside, and returning to the three overarching issues noted above, the important question is the sensibility of his substantive policy proposals, the core of which is his desire to give national governments additional regulatory flexibility. In this regard, there are a number of problems with Rodrik's approach, as well as perhaps a missed opportunity.

Turning to the key substantive points, one of Rodrik's long-standing arguments is that countries need more policy space than they are currently given under international trade rules. Rodrik says he wants ‘smart globalization, not maximum globalization’, which will be achieved by a ‘thin layer of international rules that leaves substantial room for maneuver by national governments’ (p. xix). There seem to be two main elements to the policy space argument. First, he wants developing countries to be able to use ‘industrial policy’ to promote their economic development. And, second, he wants all countries to have the ability to pursue important social policies, without interference from international trade rules.

In responding to these points, it is important to note at the outset that it is not clear there really is a lack of ‘policy space’, of either sort, under existing rules. Arguably, the rules are quite flexible. It is easy to point to specific instances where a particular government measure was found to violate WTO rules, as Rodrik does when he identifies the US–Gasoline, Japan–Alcoholic Beverages, and US–Shrimp decisions (among others) (p. 80). But a finding that a government was in violation does not necessarily mean the government did not have the policy space to achieve its goals. It may just mean that the measure in question was an instance of disguised protection, rather than a measure legitimately intended to pursue the policy in question (which most people would agree was the situation in the Gasoline case). Or it might mean that the government chose a prohibited means of doing something when there were other, permissible avenues (e.g., Rodrik seems concerned about local content requirements being prohibited, but tariffs, which are regulated but not prohibited by the WTO, could be used to achieve the same goals). While some of the cases he cites may illustrate overreaching by WTO rules, others are fairly straightforward applications of the non-discrimination principle.

With regard to the two kinds of policies at issue, I turn first to industrial policy, which he advocates as a way for developing countries to achieve economic growth and industrialization. Rodrik takes the view that economic development results from active government intervention, in particular in the form of protectionism, in the economy. This is a fairly widespread view. Of course, there is also a widespread opposing view, which is that free trade is much better for economic development. This debate has been going on for hundreds of years, and is likely to continue for hundreds more. But regardless of who is right on this issue, two key points need to be made.

First, WTO rules allow many kinds of industrial policy. The WTO does not eliminate all protectionism. It merely restricts and regulates it. Under the rules, developing countries have a great deal of flexibility to pursue protectionism. This flexibility is not unlimited, but is fairly broad. To take just a few examples, there are the special rules in GATT Article XVIII for developing countries who wish to impose trade restrictions (1) as a way to address balance of payments difficulties or (2) for economic development reasons; special rules on export subsidies for developing countries, as set out in SCM Agreement Article 27.4; and, more generally, developing countries make far fewer tariff reduction commitments than rich countries do. In addition, anti-dumping and safeguards, two of the policies Rodrik mentions when discussing the GATT favorably, are still widely used. While such measures have sometimes been challenged successfully at the WTO, the vast majority of them are not even contested (Durling, Reference Durling2003). These are not development policies per se, but they illustrate how protectionism may still be used.

To illustrate his point, Rodrik says that after China's accession to the WTO, trade rules forced China to stop using traditional industrial policy measures, and switch to currency devaluation as a means to promote its domestic economy. In Rodrik's view, the trading system would be better off if China had more flexibility to use the old methods (pp. 276–277). But examining China's tariffs, the average applied tariff is above 10% for the majority of product categories (WTO Tariff Profile). And when looking at individual tariff lines, higher tariffs are immediately apparent for a number of products (WTO Tariff Data). Thus, China does seem to have considerable flexibility to use industrial policy under the existing rules.

Second, when talking about development, Rodrik seems to take the view that this is a developing versus developed country issue. In his view, developed countries should stand back and let the developing countries do whatever they want in this regard. In response, there is a political point to be made here, which is that developed countries are unlikely to do this for the more advanced developing countries, such as Brazil and China. But regardless, there is also the problem of competition among developing countries. Let's say that the rules allowed unlimited subsidies by governments of developing countries, as Rodrik seems to suggest they should (p. 256). If that were the case, would that not lead to subsidy wars between developing countries? If China and India both were to use subsidies to develop their automobile industries, there would likely be substantial trade conflict. The point of the trade regime is to establish an agreed set of rules to help mitigate this conflict, for example by putting constraints on subsidies.

I now turn to the issue of social policies. Another way to think about the ‘policy space’ that governments have is to talk about the scope of the trade regime and how trade rules affect domestic policy-making. The scope will affect how much policy space there is. Arguably, Rodrik identifies a fundamental problem with trade rules when he says: ‘the absence of a clear and bright line between where domestic prerogatives end and external obligations begin creates substantial friction. In fact, the trade regime's growing legitimacy crisis can be traced back to this fundamental ambiguity’ (p. 81). But it is not clear that his proposals help the matter. As discussed below, even if the principles he would use to define the scope of the trade regime do establish clear boundaries, and it is not clear that they do, his specific proposals may not achieve his goals as well as some possible alternatives might.

One of his core proposals, which he has set out elsewhere in more detail (Rodrik, Reference Rodrik2007), is the expanded safeguard mechanism. Under this mechanism, after a domestic procedure to ensure the legitimacy of the objectives pursued has been undertaken, governments could impose trade restrictions for various social purposes. So, for example, he notes that ‘[a] labor union may win protection when its members are forced to compete against workers abroad who toil in blatantly exploitative conditions’ (p. 255).

But query whether trade restrictions are really the best approach here. For example, is a 20% tariff on foreign goods really the best way to protect domestic labor standards? Rodrik can accept such measures if they are the result of a satisfactory domestic process. However, trading partners may not be mollified by this. A 20% tariff looks suspicious in part because regardless of the purported goals, there is a protectionist effect. Such an approach seems inherently like disguised protectionism, despite the other policy goals that have been put forward as part of the domestic procedure.

What is surprising about Rodrik's approach is that he seems to miss a somewhat obvious alternative method of defining the scope of the trade regime so as to rein in the impact on domestic policy-making. Another way to accomplish much of what he wants to achieve is to focus trade rules on non-discrimination. There is much debate about how to define non-discrimination (Ehring, Reference Ehring2002; Hudec, Reference Hudec1998; Porges and Trachtman, Reference Porges and Trachtman2003), but however it is ultimately defined, a non-discrimination principle is one way to establish the scope of the regime. Basically, this kind of regime would emphasize the progressive liberalization of discriminatory trade measures, while continuing to allow some such measures that are considered necessary for development or other social purposes. Discriminatory measures that were permitted would be those taken through the transparent mechanism of tariffs, rather than those hidden away in hard to discover places in domestic laws and regulations.

To take a concrete example, recall Rodrik's suggestion that ‘[a] labor union may win protection when its members are forced to compete against workers abroad who toil in blatantly exploitative conditions’ (p. 255). The problem, as noted above, is that this kind of action is inherently contentious and makes trading partners suspicious. A 20% tariff suggests protection just as much as it does concerns about child labor. An alternative here would be a non-discriminatory regulation that prohibits or labels products based on their production process, including the type of labor used. This would still be contentious, of course, but arguably it would be more defensible and justifiable than a 20% tariff. And, more importantly, if implemented properly, it could, in theory, be done in a manner consistent with the current rules, with no changes to the rules necessary.

Another example Rodrik puts forward on possible US responses to Indian child labor also illustrates the missed opportunity related to the role of the non-discrimination principle. He discusses the difference between (1) sanctions to promote change to Indian policies and (2) blocking imports from India because such imports constitute ‘unfair competition’ (pp. 242–243). The latter, he says, is defensible, whereas the former is not. What Rodrik leaves out here is a fuller picture of this story. The United States already has domestic regulations on child labor. As a result, blocking imports is not quite what it sounds. Looked at as a whole, the US regulation of this issue could be seen as a non-discriminatory regulation of a particular aspect of production. Again, this is not to suggest that such regulation would not be contentious. Regulations of this sort have been challenged at the GATT/WTO on a number of occasions. But a strong argument can be made that there is a fundamental difference, in legal terms, between slapping tariffs on imports and applying a non-discriminatory regulation, and there is a good chance that the latter would not be found in violation of the rules, if done properly. It is not clear why he does not look to a non-discrimination standard as a possible resolution to his concerns.

The result of the non-discrimination approach is that policy space would be expanded considerably. Governments could pursue other policies to whatever extent they wanted, as long as the measures used were not discriminatory. This should alleviate the concerns expressed by Rodrik in many of his examples, such as the Hormones and Biotech cases. It is easy to criticize WTO rules that require health measures to be ‘based on [a] scientific risk assessment’, as Rodrik does (p. 80). But why not replace such rules with a non-discrimination requirement? As Rodrik later notes, the Hormones and Biotech measures were not discriminatory (pp. 196–197). Replacing the scientific basis requirement with a non-discrimination rule might be a better way to meet his objectives than is his safeguard tariff proposal.

In addition, such an approach makes his trilemma – involving a tension between globalization, democracy, and national self-determination – seem less problematic. A non-discrimination principle can promote significant globalization without interfering much in national, democratic affairs.

At one point, Rodrik acknowledges the point about non-discrimination, noting that the GATT regime allowed countries ‘to dance to their own tune as long as they removed a number of border restrictions on trade and generally treated all their trade partners equally’ (p. 204). But this insight seems to get lost in the shuffle.

Along these same lines, Rodrik misses an opportunity to discuss some issues that one would assume cause him great concern: investor–state dispute settlement and the TRIPS Agreement. These issues are only mentioned briefly, and never really developed. The problems with investor–state are much broader than the expropriation rules he mentions (p. 197). There are also rules on the ‘minimum standard of treatment’, performance requirements, and financial transfers. If he does not like the constraints of the WTO, one would assume that he would be apoplectic about investor–state cases, which can be brought directly by corporations without having to go through the government. And the TRIPS Agreement is more than just the rules on patents and copyright he mentions (pp. 78, 199). It would have been interesting to see Rodrik explore why intellectual property rules are now included in the trade regime.

A final point relates to Rodrik's treatment of the issues of democracy, sovereignty, and globalization. When reading this book, it is difficult not to think of Claude Barfield's Free Trade, Democracy, and Sovereignty (Barfield, Reference Barfield2001). Like Barfield, Rodrik is concerned about the constraints imposed by binding international trade rules that have expanded beyond the traditional GATT subject areas, as well as the new, more effective dispute settlement system of the WTO. Unlike Barfield, though, Rodrik does not place much emphasis on ‘sovereignty’, and he seems to avoid the term for the most part, often using substitutes. For example, he says that ‘democracy and national determination should trump hyperglobalization’ (p. xix) and also refers to ‘national self-determination’ (p. 200). Regardless of the terms used, though, the fundamental concern seems to be the same. The difference is that Barfield did not want to use that sovereignty/national determination to engage in industrial policy or trade restrictions, as Rodrik does.

With regard to democracy, Rodrik explains that democratic governance is important and in his chapter on principles he suggests that democracies should have an easier time justifying their restrictive measures through a proper domestic political process than would non-democracies (pp. 244–246). But there is an obvious problem here which he does not acknowledge: China is not going to like the worse treatment of non-democracies, and it may not be possible to have a trade regime without China. Rodrik is certainly not alone in arguing for different treatment of democracies and non-democracies. But in terms of coming up with a practical solution to the problems with the current trade regime, this is a major stumbling block.

Ultimately, in my view, Rodrik's contributions to the globalization debate are helpful, even for those who support most aspects of the current system. He puts forward a thoughtful critique and offers concrete proposals for reform. It is good to be able to question one's long-held beliefs, and Rodrik forces defenders of the regime to do that. While his specific proposals may be impractical, he has identified some real problems. In the end, these kinds of critiques of the trade regime should put it on a firmer footing and make it stronger.