The unconstrained ability to trade across national borders is known as trade.

Policy Reform and Commitment

Trade policy reforms – and other sorts of reform – are often hampered by the expectation that they may be reversed. Adjusting to reform typically involves investments, but these investments will not be made unless investors are confident that the reform will persist. These problems are mitigated if a country has a ‘commitment mechanism’ guaranteeing that the reform will be durable, and membership of a RIA can, in some circumstances, provide such a mechanism (Fernandez and Portes, 1998).

The commitment mechanism operates most obviously for trade policy – membership requires that tariffs with member countries be cut, and reneging on agreed internal liberalization is likely to bring swift retaliation by partner countries. However, it has been argued that RIAs are valuable as commitment mechanisms for a much wider range of measures. Although NAFTA was ostensibly about trade policy, an important part of its motivation was the desire on the part of both the Mexican and US governments to lock in the broad range of economic reforms that the Mexican government had undertaken in the preceding years. The EU Articles of Agreement with eastern European accession candidates are explicit in promoting ‘full integration into the community of democratic nations.’ And the intervention of other Mercosur countries is credited with having averted a military coup in Paraguay in 1996 (Survey on Mercosur, The Economist, 12 October 1996). Paradoxically, it is even suggested that the value of a RIA as a commitment mechanism is greatest in areas other than trade policy, because there is already a way committing to tariff reductions – the tariff bindings of the GATT/WTO.

13.3.1 Brief history of distortions to agricultural incentives

Some agricultural and other trade policy developments over the past half century or so have happened quite suddenly, been unpredicted, and been transformational, but most have been more gradual. For decades, agricultural protection and subsidies in high-income (and some middle-income) countries have been depressing international prices of farm products. The Haberler (1958) report to GATT Contracting Parties forewarned that such distortions might worsen, and indeed they did between the 1950s and the early 1980s (Anderson and Hayami, 1986). Meanwhile, the governments of many developing countries directly taxed their farmers, overvalued their currency and pursued import-substituting industrialization by restricting imports of manufactures. Together, those measures strongly discouraged agricultural production in developing countries (Krueger et al., 1988, 1991). Since the 1980s those disincentives have been easing, however, as revealed in the new database of agricultural distortions over the past half-century compiled recently by the World Bank (Anderson and Valenzuela, 2008). That database indicates the extent to which government-imposed distortions created a gap between domestic producer prices and what they would be under free markets, known as the NRA.24 Since the 1980s, some high-income country governments have also provided so-called “decoupled” assistance to farmers. As that support in principle does not distort resource allocation like direct price supports, its NRA has been computed separately and is not included for comparison with the NRAs for other sectors or for developing countries.

Each farm industry is classified either as import-competing, or a producer of exportables, or as producing a non-tradable (with its status sometimes changing over the years), so as to generate for each year the weighted average NRAs for the two different groups of covered tradable farm products. Also generated is a production-weighted average NRA for non-agricultural tradables, for comparison with that for agricultural tradables via the calculation of a relative rate of assistance (RRA), defined in percentage terms as:

(13.1)RRA=100[100+NRAagt100+NRAnonagt−1],

where NRAagt and NRAnonagt are the percentage NRAs for the tradable parts of the agricultural (including non-covered) and non-agricultural sectors, respectively. Since the NRA cannot be less than −100% if producers are to earn anything, neither can the RRA (since the weighted average NRAnonagt is non-negative in all the country case studies). If both of those sectors are equally assisted, the RRA is zero.

Historically, national NRAs to agriculture have tended to be higher, the higher a country’s income per capita and the weaker a country’s agricultural comparative advantage. There has also been a (somewhat weaker) tendency since the 1960s for manufacturing protection to be lower, the higher a country’s income per capita and the stronger a country’s manufacturing comparative advantage. Together these tendencies would expect one to observe the RRA to farmers to be positively correlated with per capita income and negatively correlated with an index of comparative advantage in farm products. This is indeed what the DAI database reveals (Figure 13.1). Figure 13.2 shows the RRA has been rising over time for developing countries as a group and also for high-income countries prior to the 1990s. The developing countries’ RRA rose from around −50% in the latter 1960s to almost zero in 2000–2004, while the RRA for high-income countries rose from 14% in the latter 1950s to a peak of just above 50% in the late 1980s. A movement in the RRA towards (away from) zero might indicate an improvement (worsening) in economic welfare, suggesting that the welfare cost of developing country policies may have been falling but may begin increasing if they follow the high-income countries’ earlier example in raising their now-positive average RRA further. That is certainly what Korea and Taiwan did in following Japan, and China, India and other developing countries appear to be on a similar trajectory (Anderson, 2009b).

A disaggregation of the NRA estimate for the agricultural sector into the NRAs for the export and import-competing subsectors, as in Figure 13.3, reveals that developing country exporters of farm products faced a tax of around 50% on average in the first decades of post-colonial government; but that the average rate of taxation has gradually fallen to almost zero since the mid-1980s. Meanwhile, however, the NRA for import-competing farmers in developing countries has been positive and steadily rising throughout this period (apart from a spike in the mid-1980s when international prices fell to a near-record low as a consequence of a farm export subsidy war between the two sides of the North Atlantic). The trend for exporters could have reduced the welfare cost of agricultural distortions in developing countries, but the fact that import-competing farmers were increasingly assisted reduces that possibility. As for high-income countries, Figure 13.3(b) shows that their exporters received rising support until the end of the North Atlantic farm export subsidy war, but that import-competing farmers enjoyed higher and faster-rising support over that period than exporters.

Regional trade agreements: taxonomy

In the dictionary of trade policy terms, regionalism is described as ‘actions by governments to liberalize or facilitate trade on a regional basis, sometimes through free-trade areas or customs unions’. Regionalism refers to formal economic cooperation and economic arrangements of a group of countries aimed at facilitating or enhancing regional integration. For instance, a group of countries enter into an RTA to facilitate economic integration.

RTAs can take various forms: a preferential trade area, a free trade area, a customs union, a common market and economic union. In a preferential trade area, trading partners grant partial preferential tariff reductions to each other. In a free trade area, members eliminate all tariffs and non-tariff barriers among themselves, but each member can set its own tariff rates on non-members. A customs union is a free trade area, but members adopt a common external tariff on non-members. A common market goes beyond a customs union by allowing free movements of factors of production. Finally, economic union involves integrating national economic policies, like fiscal and monetary policies. The higher the level of regional integration, the greater the complexity, as shown in Figure 1.2. RTAs differ in configuration. They may be bilateral (an agreement between two parties) or plurilateral (an agreement among three or more parties). More complex agreements occur when one (or more) of the parties is an RTA itself or all parties are themselves distinct RTAs. RTAs also differ in scope. The simplest form is the exchange of preferences on a limited number of products among the parties. The more complex ones go beyond tariff elimination to include services, investment, competition policy, government procurement, intellectual property rights, etc.

Until December 2007 the most common category was FTAs, which account for 82 per cent of all RTAs notified and in force, while partial-scope agreements and customs unions (CUs) each account for 9 per cent. Of the projected RTAs, 93 per cent are intended to be FTAs, 6 per cent are partial-scope agreements and only 1 per cent are customs unions (Fiorentino et al., 2008). The dominance of FTAs over customs unions is probably due to the fact that they are faster to conclude and require a lower degree of policy coordination among the parties, since in an FTA each party maintains its own trade policy vis-à-vis third parties. In contrast, customs unions require the establishment of a common external tariff and harmonization of external trade policies, implying a greater loss of autonomy over parties’ commercial policies and longer and more complex negotiations and implementation periods. Furthermore, the majority of FTAs are concerned with strategic market access, often unbounded by geographical considerations; in customs unions, on the other hand, geographical considerations play a pivotal role in defining the objective of economic (and often political) integration among the parties concerned (Crawford and Fiorentino, 2005). As for partial-scope agreements, their limited trade coverage, poor implementation record and low visibility make them much less attractive to countries, including developing ones, which are committed to comprehensive trade liberalization. The configuration of RTAs is varied and gradually becoming more complex, with overlapping RTAs and networks spanning within and across continents at regional and subregional levels. Countries are opting for a simple two-party RTA configuration rather than more troublesome plurilateral RTAs, which are more common in customs unions. Bilateral agreements account for over 75 per cent of all RTAs notified and in force, and for almost 90 per cent of those under negotiation. In terms of their scope and depth, RTAs differ considerably, with some providing for the exchange of tariff preferences on a limited range of products and others being highly comprehensive in coverage and including wide-ranging trade regulatory regimes. Given the requirements prescribed by the WTO provisions on RTAs, partial-scope agreements fall under the legal cover of the ‘enabling clause’, concern exclusively agreements among developing countries and in most cases tend to have limited product coverage. FTAs and customs unions falling under the legal cover of GATT Article XXIV and/or Article V for trade in services are comprehensive in scope, and especially the most recent agreements often go beyond the WTO regulatory framework to include provisions on investment, competition, intellectual property, environment and labor among others.

Traditionally, RTA formation has occurred between so-called ‘natural’ trading partners: geographically contiguous countries with already well-established trading patterns. Australia and New Zealand, NAFTA countries, the EC (European Community), EFTA (European Free Trade Association) and CEFTA (Central European Free Trade Agreement) provide good examples. Indeed, most countries sign their first RTA with one or several neighboring or regional partners. Southeast Asian countries’ participation in ASEAN, sub-Saharan African groupings such as CEMAC (Economic and Monetary Community of Central Africa) and SACU and the western hemisphere groupings of CARICOM (Caribbean Community), the CACM (Central American Common Market) and Mercosur are all prime examples. However, once a country has exhausted its regional prospects, it may begin to look further afield for preferential partners. This trend is most evident in countries of the western hemisphere, Europe and increasingly the Asia-Pacific (Karim, 2006). Frankel et al. (1998) argue that geographical proximity plays a significant role in the success of regional trade.

2.4 Trends in Modern PTAs

We provided a snapshot of PTAs in 2011. In the Online Appendix II, we graph the evolution of the share of PTAs with different provisions since 1991. Here we point to a few key points. In terms of depth the most important trend is the increased prevalence of provisions addressing contingent protection, product standards, and public procurement. In terms of economic breadth, since 2000 there has been an increase in service provisions and labor market regulations. Finally, in terms of noneconomic areas we see increases in environmental laws from less than 30% to almost 50% and also in human rights and illicit drugs.

3.1.4 Reciprocity and MFN

The principles of reciprocity and MFN are two major pillars of the GATT-WTO. The role of these principles has been examined by Bagwell and Staiger in a number of writings, and most notably in the context of the perfect-competition setting of Bagwell and Staiger (1999a). I will start by focusing on the principle of reciprocity in the context of a two-country model (where MFN of course plays no role).

Bagwell and Staiger define a reciprocal change in trade policies as one that leaves world prices unchanged.55 More specifically, they distinguish between two notions of reciprocity. The first one is a principle guiding tariff negotiations starting from the Nash equilibrium: Bagwell and Staiger show that, if tariff negotiations satisfy reciprocity (that is, if they leave world prices unchanged relative to the Nash equilibrium), they will lead to a Pareto-improvement over the Nash equilibrium, although in general they will not lead all the way to the Pareto-efficient frontier.56 An important note of interpretation is the following: in the GATT-WTO the principle of reciprocity as it applies to tariff negotiations is not a strict rule, but just an informal principle, so Bagwell and Staiger’s (1999a) analysis of reciprocity-guided negotiations is best interpreted in normative terms, as highlighting how the negotiation outcome would be affected if reciprocity were strictly imposed as a rule.

The second notion of reciprocity considered by Bagwell and Staiger applies to tariff renegotiations. Unlike the informal reciprocity principle that applies to negotiations, GATT-WTO does impose a formal rule of reciprocity for tariff renegotiations, specifically in Article XXVIII of GATT. Bagwell and Staiger formalize this rule through a two-stage negotiation game, as follows: (i) in the first stage, governments negotiate a pair of tariffs, with the disagreement point given by the Nash equilibrium tariffs; (ii) in the second stage, governments can renegotiate the tariffs, and if the renegotiation fails, a government can unilaterally change its tariff from the level that was agreed upon in the first stage, but at the condition that the trading partner get compensated through a reciprocal tariff change.

Before proceeding, I make an observation that will be useful later on. Notice the nature of the disagreement point for the process of renegotiation under reciprocity: if the renegotiation fails, a government can choose to unilaterally “breach” the contract and compensate the trading partner, with the compensation taking the form of a reciprocal tariff increase by the trading partner. This observation leads me to note that there is a simpler way to intepret such reciprocity rule, which I find more illuminating. Rather than imposing a constraint on renegotiation, this rule can equivalently be modeled as changing the nature of the contract itself, from one that specifies tariff commitments without allowing for “breach” (in law and economics jargon, a “property” contract), to one that allows a government to breach the contract in exchange for a certain amount of “damages” (in law and economics jargon, a “liability” contract). Notice that, if reciprocity is modeled in this way, simply as a liability rule incorporated in the initial contract, then there is no scope for renegotiation in Bagwell and Staiger’s setting, and we can simply think of governments as negotiating a contract within this particular class (liability contracts with breach damages given by reciprocal tariff changes), ignoring renegotiation.57

Bagwell and Staiger’s main results regarding the rule of reciprocity are two. The first one is that the only efficient tariff pair that can be implemented under reciprocity (or, adopting my interpretation above, when the contract is restricted to be a liability contract with breach damages given by reciprocal tariff changes) is given by the politically efficient (PO) tariffs. This result is a consequence of the fact that, when viewed in tariff space, each government’s iso-payoff curve is tangent to the iso-world-price curve at the PO point, thus starting from this point there is no unilateral incentive for a government to move along the iso-world-price curve. This in turn implies that the reciprocity-constrained Pareto frontier lies inside the unconstrained Pareto frontier, except at the PO point. The second result is that, when governments bargain over the reciprocity-constrained Pareto frontier, the outcome is generically inefficient (it is not the PO point), but it tends to be closer to the PO point as compared with the unrestricted bargaining scenario.58

Bagwell and Staiger interpret these results as suggesting that the reciprocity rule induces a rebalancing of power across countries, since it moves the negotiation outcome toward a point that is not affected by bargaining powers (the PO point). This conclusion resonates with the GATT’s emphasis on “rules” vs “power,” but as remarked above, the rebalancing of power induced by the reciprocity rule in general entails an “efficiency penalty.” Bagwell and Staiger then argue informally that reciprocity may provide efficiency gains if it encourages weaker countries to participate in GATT. This idea is based on McLaren’s (1997) model: recall from Section 2.2 that in this model a small country trading with a large country may prefer to stay out of a TA to avoid being “held up,” and for this reason the large country would like to commit not to exploit its bargaining power, in order to encourage the small country to participate in the TA. Thus the broad idea is that, if one thinks of GATT as initially including a set of large/powerful countries, but there is also a set of smaller/weaker countries that may consider accessing GATT at a later stage, the initial members may prefer to commit not to exploit their bargaining power in future negotiations, in order to encourage other countries to seek participation.

I will make two further comments about Bagwell and Staiger’s analysis of the reciprocity rule. The first one is a “devil’s advocate” comment. Bagwell and Staiger’s interpretation of the results outlined above is that reciprocity “works well” when TOT externalities are the only motive for a TA. But one might argue that the model tells the opposite story: reciprocity causes the negotiation outcome to be inside the Pareto frontier, so the world would be more efficient without reciprocity. A more cautious interpretation of Bagwell and Staiger’s analysis would be that it provides a positive (as opposed to normative) evaluation of the reciprocity rule, highlighting that this rule has a distributional effect (which might be desirable in a richer model that includes participation considerations) and an efficiency cost.

The second comment is that, aside from the participation argument outlined above, Bagwell and Staiger’s analysis is suggestive of another potential efficiency benefit of reciprocity. Since, as I argued above, imposing the reciprocity rule is akin to structuring the TA as a (specific type of) “liability” contract, this might be appealing as a way to inject flexibility in the TA when countries are subject to shocks and the TA cannot be fully contingent. But this is just suggestive, since there is no uncertainty or contract incompleteness in the model.59 This brings me back to one of my overarching points, namely that a better understanding of TA design requires bringing transaction costs explicitly into the picture.

After their analysis of reciprocity within a two-country setting, Bagwell and Staiger (1999a) turn to an examination of the implications of the MFN rule within a multi-country setting. As I mentioned in Section 2.1.2, in a multilateral world where governments can set discriminatory tariffs, there is a whole vector of bilateral offshore prices, which generates a complicated pattern of international policy externalities, but the MFN rule has the effect of channeling all these externalities into a single world-price externality. Building on this observation, Bagwell and Staiger show that the MFN rule, if used in tandem with the reciprocity rule, guides countries toward the PO tariffs (which, recall, are efficient under MFN). The key point here is that the combination of MFN and reciprocity has similar effects in a multi-country world as the reciprocity rule does in a two-country world.

So far I have focused on the implications of reciprocity and MFN within a perfectly competitive environment. It has been argued, however, that reciprocity and MFN can be rationalized also within an imperfectly competitive environment, where the motives for a TA go beyond the correction of TOT externalities. In particular, Ossa (2011) argues that, in a monopolistic-competition setting, reciprocity and MFN can be interpreted as helping countries internalize the production-delocation externalities generated by trade policies.60

8.5.2 Distributional impact

We next report the impacts of the cap-and-trade policies on household welfare, as given by the equivalent variation in full wealth in Equation (8.101). Recall that the equivalent variation in full wealth is the wealth required to attain the welfare associated with a new policy at base-case prices, less the wealth required to attain the base-case welfare at these same prices. We consider equivalent variations for each of the 244 household types, cross-classified by the demographic categories presented in Table 8.3.26

In Chapter 17 we describe the wide range of full expenditure values for each demographic category. Figure 8.9 shows the welfare effects for the three policy scenarios at the mean full wealth for each of the 244 household types. The percentage losses are smaller than those for GDP because households respond to the changing prices by changing their consumption patterns. The welfare losses also are smaller because of the offsetting increases in leisure demand and the lump-sum transfers of allowance revenues.

The households with lower full wealth suffer a larger percentage loss in welfare. Smaller households and others with lower full wealth consume less leisure and have larger budget shares of consumer goods. Hence, they are more adversely affected by the direct and indirect effects of mitigation policy. Moreover, as emissions targets are tightened, the welfare losses increase at an increasing rate as mean household wealth decreases. Lower expenditure households are harmed more and, the more aggressive the abatement policy, the more they are harmed.

Figures 8.10 and 8.11 show the welfare effects associated with only the 25% target. In Figure 8.10, the 244 household types are arranged from the lowest to the highest levels of mean full wealth. In Figure 8.11, the 244 household types are arranged from the most to the least adversely affected. The principal curve is the solid line in each graph labeled “At mean wealth.” This shows the welfare impact on households with the mean wealth among those with the same demographic characteristics. For mean wealth, all households experience a welfare loss ranging from –0.02 to –0.22% of full wealth are included. The most negatively affected households consist of one child with one adult living in the rural South and headed by non-white females. The least negatively affected are large urban households in the West, households with three or more children and three or more adults headed by non-white males.

To illustrate how the policy affects households with different levels of full wealth, the effects in Figure 8.10 and 8.11 also are shown for half and twice mean wealth. The population-weighted average welfare effects are –0.17, –0.11 and –0.05% of lifetime expenditure at the half mean, mean, and twice mean levels, respectively, as represented by the horizontal lines in Figure 8.11. It must be emphasized that these are the average household effects. Figures 8.10 and 8.11 show that the effects of the policy change are regressive; the equivalent variations become more negative as full wealth decreases both across and within demographic groups. However, it should be noted that in all cases the welfare losses are relatively small, the worst case being under 0.3% of full wealth.

Figure 8.12 decomposes the welfare effect by isolating the impact of price changes alone. The solid line in Figure 8.12 shows the solid curves from Figures 8.9 and 8.10 for mean wealth. The dashed line below it shows the welfare effects due solely to price changes, holding household full expenditure at its base case value. In the absence of changes in expenditure, households experience net welfare losses in the range of 0.23–0.43% of their full wealth. However, the lump-sum redistributions required to hold real government spending at its base-case levels partially compensate the price effects.

With leisure such a large share of household budgets, there is a natural concern as to the influence of family size on the findings summarized above. Accordingly, we examine the two most populous segments of the household sample – households with two adults and no children (28.8% of all households), and households with one adult and no children (29.5%). The dominance of leisure in full wealth is evident here. In Figures 8.9, 8.10 and 8.12, single-person adult households are concentrated at the lower end of the spectrum for lifetime expenditure while two-person adult households are concentrated in its middle range. However, what matters most here are the robust findings of regressivity when controlling for leisure this manner. For both groups, climate policy results in larger welfare losses at lower levels of full wealth and, in both cases, this pattern is invariant to scale.

Table 8.14 decomposes the welfare losses across the demographic details of the 244 household types. These are summarized by the population-weighted averages within each group. Clearly, households containing three or more adults are generally better off than those with two adults, which in turn are better off than single-adult households. This is not surprising in that larger households generally are wealthier in terms of their lifetime spending on goods, services and leisure. Within these groups, the presence of children is of equal interest. Households with three or more adults are better off with three or more or no children and are worse off with one or two children. Households with two adults fare progressively worse the fewer children they have whereas the opposite occurs in households with only one adult. Among single-adult households and within each grouping based on the number of children, rural households headed by females fare worst.

Table 8.14. Household welfare effects, 25% target (population weighted-average equivalent variations as a % of full wealth)

Full wealthHalf meanMeanTwice meanChildren, adults per household 3+, 3+–0.124–0.064–0.004 2, 3+–0.127–0.067–0.007 1, 3+–0.125–0.065–0.005 0, 3+–0.124–0.064–0.004 3+, 2–0.144–0.084–0.023 2, 2–0.146–0.086–0.026 1, 2–0.149–0.089–0.029 0, 2–0.152–0.092–0.032 3+, 1–0.222–0.162–0.102 2, 1–0.216–0.156–0.096 1, 1–0.213–0.153–0.093 0, 1–0.213–0.153–0.093Region of household Northeast–0.165–0.105–0.045 Midwest–0.176–0.116–0.056 South–0.173–0.113–0.053 West–0.156–0.096–0.036Race and gender of household head Non-white female–0.202–0.142–0.082 White female–0.201–0.141–0.081 Non-white male–0.160–0.100–0.040 White male–0.154–0.094–0.033Location of household Urban–0.166–0.106–0.046 Rural–0.193–0.133–0.073 Overall–0.168–0.108–0.048

In the sample, 18.9% of the household population resides in the Northeast, with 23.0, 36.5 and 21.6% residing in the Midwest, South and West, respectively. Most of the households with large welfare losses are located in the South or Midwest and the largest losses occur in the South. The households with the smallest proportional losses are in the West and, on average, this region fares the best followed by the Northeast, South and Midwest.

Households headed by non-white females comprise 7.4% of the sample population. Households headed by white females comprise 22.5% of the sample. Households headed by non-white and white males, comprise 10.3 and 59.8% of the sample, respectively. The household types with largest welfare losses are headed by females though, on average, there is not much difference between those headed by whites or non-whites. Male-headed households fare much better, owing to their greater wealth. Here, again, the average difference between the races is not large. Overall, the welfare gap across the sexes is much more significant than that across the races. The households with the largest losses are concentrated in rural areas. The larger (92.1%), wealthier urban population fares better than the smaller (7.9%), poorer rural segment.

1 Introduction

This chapter surveys the broad features of and developments in the use of trade policy across countries, within countries, and over time. Our goal is to describe and, whenever possible, quantify the extent of cross-sectional heterogeneity in applied commercial policy across countries, their economic sectors, and their trading partners, over time.a We construct a relatively comprehensive picture of trade policy for 31 economies that represented 83% of the world's population and 91% of the world's GDP in 2013. Our main conclusion is that substantial trade policy barriers remain as an important feature of the world economy today.

As a starting point, we must define what we mean by a trade policy “barrier.” As we will illustrate, there is an ever-increasing set of policy instruments that both researchers and trade negotiators would like to better understand because they are perceived to have a trade-restricting effect. Until recently, researchers have had little information about border policy instruments. Indeed, as Anderson and van Wincoop noted: “[t]he grossly incomplete and inaccurate information on policy barriers available to researchers is a scandal and a puzzle” (Anderson and van Wincoop, 2004, p. 693). The good news is that research no longer needs to operate behind a veil of ignorance created by the lack of information on border barriers; product-level data on many of the important trade policy instruments imposed at the border are now routinely being made available for most economies in the world.b

The picture of the empirical landscape of trade policy that emerges is complex, with substantial variation along some dimensions and little variation along others. We frame our assessment of this trade policy variation by centering our analysis around five main questions:

1.

Do some countries have more liberal trading regimes than others?

Within countries, which industries receive the most import protection?

How do trade policies change over time?

Do countries discriminate across their trading partners when setting trade policy?

How liberalized is world trade?

As we examine each question, we document heterogeneity in trade policy arising along important dimensions.c Whenever possible, we also comment on the factors that are understood as driving this variation. Here, we offer five brief answers, with a promise to expand on them in the coming pages.

Question 1: Do some countries have more liberal trading regimes than others?

Yes, there are large differences across countries in the average level of import tariffs. High-income countries have more liberal regimes than middle income countries which, in turn, have more liberal regimes than low income countries. To this broad picture we add two more subtle observations. High-income countries appear even more open when we expand the set of policies to include preferential tariffs on imports that they offer to selected trading partners. On the other hand, both high-income and emerging economies appear less open when we expand the set of policies beyond tariffs to include policies such as temporary trade barriers and quantitative restrictions.

Question 2: Within countries, which industries receive the most import protection?

With respect to import tariffs, agricultural products and foodstuffs are protected almost everywhere, regardless of a country's level of development. Textiles, apparel and footwear are more protected than other manufactured goods. Minerals and fuels tend to face few import barriers. Furthermore, import tariffs on final goods are higher than those on intermediates in all sectors, everywhere. When expanding the policy set beyond tariffs, however, the sector-level variation in import protection levels becomes much more muddled.

Question 3: How do trade policies change over time? Are countries consistently liberalizing, or are there reversals?

Over the last 20 years, tariffs have been trending down around the world. Among high-income countries, changes in applied tariffs are mostly stable around this trend; there is a bit more fluctuation for lower income countries. However, this broad trend is partially deceiving for many high income and emerging economies; they have switched from tariffs toward antidumping and safeguards policies to implement higher frequency changes to their levels of import protection.

Question 4: Do countries discriminate across their trading partners when setting trade policy? If so, by how much?

The short answer to the first part of this question is yes. Many countries discriminate across trading partners by offering lower levels of import protection—ie, lower applied tariffs that provide more favorable market access to preferred trading partners relative to the nondiscriminatory tariffs offered under multilateral agreements. On the other hand, many countries discriminate across trading partners by imposing higher levels of import protection—ie, antidumping import restrictions designed to limit the market access of particular trading partners, including new entrants such as China. Answering the second part of this question is much more difficult.

Question 5: How liberalized is world trade?

In historical context, the answer is probably a lot relative to previous eras. However, it is much less liberalized than many economists probably realize, especially when taking into account border barriers beyond tariffs. Finally, very little is known about how “liberalized” world trade is once we expand the analysis and consider potential trade-restricting effects of behind-the-border policies.

In our attempts to answer these five questions, three important themes emerge.

First, trade is restricted through the use of many different policy tools. The fundamental dichotomy in the lexicon of import policies has been between price-based measures (import tariffs) and quantity-based measures (quotas). However, a variety of specialized categories have arisen within these broad classes. The development of trade agreements has played an important role in both constraining access to certain policies, and yet also making other policies more readily available under certain types of legal-economic conditions. The result is an extensive variety of policies in use at any moment in time. Moreover, across countries and sectors, the trade regime can exhibit extensive heterogeneity in the level of restrictiveness. Including all border policy barriers is thus likely to be important; for example, in their major contribution to estimating the combined restrictiveness of various trade policies using data from the late 1990s and early 2000s, Kee et al. (2009) conclude that restrictiveness measures that include nontariff barriers are 87% higher on average than measures based on tariffs alone.

A second emerging theme is that history tends to repeat itself. While some of the most popular forms of border policies have changed over the decades since World War II, some of the same sectors remain protected or have repeated episodes of protection. Furthermore, the circumstances in which countries raise their barriers to trade resurface time and again.

The third theme is that the reduction of traditional border barriers and integration of economic activity, even though arguably incomplete, has opened up new areas of policy conflict that are expected to grow in importance. Bilateral frictions between trading partners have moved beyond tariffs and quotas to the international externalities associated with domestic policies—ie, domestic tax and subsidy regimes, health and safety standards for products, as well as labor and environmental regulations. In order to survey the empirical landscape of these “behind-the-border” policies that potentially restrict trade, we use case studies to highlight important themes. Partly because of the lack of internationally comparable data on domestic policies, rigorous empirical work in this area is very thin. While this area is of increasing importance, the literature is unsettled as to the positive and normative understanding of the extent to which regulatory policies unduly inhibit trade and what, if anything, trade agreements could or should do about it.

As we catalogue and describe what is known about trade policy in 2013 and 2014—the most recent years for which we observe near-complete data reporting—one objective is to correct the widely-held misunderstanding that trade is already free from policy impediments. Among academic economists, there remains considerable disagreement about the quantitative importance of different types of trade barriers. While firm-level studies infer that sizeable additional costs associated with international commerce relative to domestic commerce must exist,d an open question is whether these important trade costs are due to trade policy or to something else. Anderson and van Wincoop (2004), through a combination of direct observation and inferences from a gravity trade model, quantify the representative border-related trade costs for an industrialized country at 44% ad valorem and the representative transportation costs at 21%. In contrast, Hummels (2007, p. 136) suggests that the importance of policy barriers has been completely eclipsed by real transportation costs: “For the median individual shipment in US imports in 2004, exporters paid $9 in transportation costs for every $1 they paid in tariff duties.”

Although we are not be able to quantify the relative contributions of different forms of trade costs, we do hope to illustrate why many summary measures of trade policy are not appropriate for understanding the full economic importance of policy barriers to trade. Three simple examples help to illustrate the point. First, the real world consists of significant nontariff policy instruments; these include a number of border policies that, while technically applied as a tariff, are typically not captured in the reported data of “headline” measures of tariffs. Second, trade policy is commonly applied in a nonuniform manner across trading partners. Third, applied trade policy can vary considerably over time, especially in response to the business cycle, movements in the real exchange rate, or due to trade volume shocks; nevertheless, these policy changes are not made through applied ad valorem import tariffs, but via some other, less transparent, policy tool.

In the coming sections, we document that policy barriers to trade still exist; they vary considerably across products, trading partners and time; they take many forms; and they arise under legal frameworks (established by international trade agreements) that can result in subtle differences in both the frequency of their use and of their trade-reducing potential.

Section 2 begins by analyzing the focal policy of the trading system—the import tariff. This section examines different dimensions of tariff data for our sample of 31 economies to provide an initial response to each of our chapter's first four questions. We introduce the most-favored-nation (MFN) ad valorem import tariff—ie, the “headline” border policy instrument that countries impose under the WTO. While applied tariff levels are relatively low in historical terms, significant cross-country heterogeneity remains; furthermore, the negative correlation between applied tariffs and income per capita is even more pronounced once we consider the countries’ “legal” commitments that are the upper limits for their tariffs. Next, we characterize cross-sectional variation in applied tariffs across sectors and inter-temporal variation across countries over 1993–2013. We then introduce which countries and sectors apply specific—or per unit—duties, and we describe what makes them distinct from ad valorem tariffs. Finally, we characterize the lower-than-MFN tariffs that countries apply under a variety of preferential trade arrangements. Because these allow countries to take on additional liberalization, we illustrate the extent to which countries apply tariffs so as to discriminate among trading partners.

In Section 3 we introduce the other major (nontariff) border barriers, and we use them to further inform our chapter's main themes.e Contemporary use of temporary trade barrier policies—such as antidumping, safeguards, and countervailing duties—by emerging economies in particular has been increasing, and these import restrictions are characterized by both heterogeneity across economic sectors and discriminatory use against particular exporting country trading partners. We then turn to a description of other nontariff policies, such as quotas, price undertakings, voluntary export restraints, and other administrative hurdles (customs valuation and import licensing) that governments can manipulate to restrict trade at the border.

Once we have described the contemporary landscape of border policies, Section 4 examines the longer-term evolution of import tariffs and other border barriers in order to put the current system into better historical context. We start with information on the evolution of tariffs from the late 1940s through the late 1990s. Moreover, we provide a brief history of the ebb and flow in the use of a number of special import restrictions over this longer time horizon. This includes import restrictions to safeguard the balance of payments; the discriminatory treatment of Japan despite its GATT accession in 1955; the multifiber arrangement (MFA) and other voluntary export restraints; the sectoral carve out for agriculture; and special and differential treatment for developing countries. This section not only clarifies how the current landscape of trade policy arose, but it also allows us to emphasize our second theme of history frequently repeating itself, albeit through different policy tools, by different countries, or against different trading partners. Finally, this section helps provide a partial answer to our fifth question, “how liberalized is world trade?” In comparing the border barriers of today with the border barriers of the past, the current system is one that appears relatively open.

Section 5 then returns to the contemporary landscape by introducing a set of “behind-the-border” policies that have the ability to substantially impact international commerce. These include domestic subsidies and taxes, as well as standards and regulations. A comprehensive characterization of such data is notoriously difficult and fraught with measurement concerns; thus we survey case studies from recent policy conflicts in order to highlight important areas. Our survey covers roughly 10% of the population of formal trade disputes arising during the WTO's first 20 years. We conclude that the next major frontier for the world trading system involves how it confronts the trilemma of respecting local tastes and objectives in domestic policy, internalizing cross-border policy spillovers that operate through trade flows, and facilitating greater trade integration to sustainably maximize the value of the world's productive resources.

Finally, Section 6 concludes with a brief discussion of how our results may also inform other areas of research in international economics beyond studies of commercial policy.

3.3 Reciprocity Beyond the Terms-of-Trade Theory

In the preceding sections we have considered reciprocity from the perspective of a variety of models that fall within the terms-of-trade theory of trade agreements, as well as a number of extensions to those models. We close this section with a brief discussion of reciprocity from modeling perspectives that fall outside the terms-of-trade theory, including environments where governments have a limited set of trade policy instruments, where offshoring is prevalent, and where governments seek to make commitments to their own private sectors through trade agreements.

A first observation is that, in each of these environments, we may continue to expect that tariff changes conforming to reciprocity as we have defined reciprocity above will hold the terms of trade fixed. This is because, as we noted above, we have derived this property of reciprocity in a modeling framework that features little more than market clearing and trade balance, and that is thus general enough to include each of the environments that we consider below.

We therefore turn to the remaining two questions around which we have organized our discussion of reciprocity above. First, beginning from their Nash tariffs, can both countries gain from reciprocal liberalization in these alternative modeling environments provided they do not go too far? And second, in these environments can the political optimum still be singled out as the only point on the efficiency frontier that is robust to renegotiation subject to reciprocity?

Consider first the case of missing trade policy instruments. For this case we return to the competitive partial equilibrium model with missing instruments described in Section 2, where we assumed that export policies are prohibited so that the home government has only an import tariff τx and the foreign government has only an import tariff τy*. In this environment, the Nash tariffs are defined by the first-order conditions

Home:Wpˆxdpˆxdτx+Wpxw∂pxw∂τx=0Foreign:Wpˆy**dpˆy*dτy*+Wpyw*∂pyw∂τy*=0.

With Wpˆx<0and Wpˆy**<0implied, an immediate observation is then that, beginning from their Nash import tariffs, both countries can gain from at least a small amount of reciprocal liberalization, because such tariff reductions lower pˆxand pˆy*without inducing any welfare-relevant changes in pxwand pywand therein raise home and foreign welfare according to Wpˆx<0and Wpˆy**<0. So our earlier answer to the first question above is unchanged by the limitation we have imposed here on the set of trade instruments. But our earlier answer to the second question is overturned in this environment. This follows directly from the result reported earlier, that with missing trade policy instruments it is no longer generally the case that politically optimal policies are internationally efficient, because there is a local price externality that now persists at the political optimum. In fact, as we argued above, when export policies are missing the political optimum will be inefficient whenever export industries enjoy political support in the objective functions of their governments.

We may therefore conclude that in an environment with missing trade policy instruments, the political optimum cannot in general be singled out as the only point on the efficiency frontier that is robust to renegotiation subject to reciprocity, because the political optimum does not itself generally rest on the efficiency frontier in this environment. Moreover, in light of the fact that any efficient point that does not correspond to the political optimum is susceptible to such renegotiation, we can make a stronger statement: in general for this missing-instrument setting, no point on the efficiency frontier is robust to renegotiation subject to reciprocity. This suggests that the attractive features of reciprocity that hold in the context of the terms-of-trade theory and that we have emphasized above are substantially diminished in environments where trade policy instruments are missing.

A key question, then, is how to interpret the possibility of missing trade policy instruments. One possible interpretation is that countries do not in fact possess complete sets of trade policy instruments, with a possible candidate for missing instruments being export subsidies whose funding needs might make these policies simply inaccessible to all but the richest countries. Under this interpretation a trade agreement must be designed to both internalize the terms-of-trade externality and provide the missing trade policy instruments. A second possible interpretation is that these trade policy instruments are not truly missing, but rather their use has been prohibited by international agreement. Under this second interpretation the problem to be solved by a trade agreement is still fundamentally the terms-of-trade problem, but as part of the approach to solving the terms-of-trade problem the agreement has altered the nature of the policy externalities with which the agreement itself must contend. From this perspective an important question is whether the increasingly stringent prohibition on the use of export subsidies as this prohibition has evolved from GATT to the WTO (see, for example, Sykes, 2005) creates such a missing-instruments environment for WTO member governments; and if so, whether the GATT/WTO traditional reliance on reciprocity is growing increasingly at odds with an institution well designed to solve the terms-of-trade driven Prisoners’ Dilemma problem.cb

Ossa (2011) provides an interesting answer to this question. He demonstrates that, in a general equilibrium monopolistic competition model of firm delocation where export policies are ruled out, an adaptation of the definition of reciprocity can be found which largely preserves the properties of reciprocity we have emphasized above. Ossa's model has both a manufacturing and a nonmanufacturing sector, with the manufacturing sector composed of many firms producing differentiated manufacturing goods and representing the sector where firm delocation effects occur. In Ossa's model the terms-of-trade manipulation problem is completely absent because, as in the partial equilibrium monopolistic competition model of firm delocation we described in Section 2, only export policies can have terms-of-trade impacts, and as we noted in Ossa's model the use of export policies is ruled out by assumption. This allows Ossa to highlight the firm-delocation problem that is then transmitted across countries though local-price channels. Ossa shows in this setting that if reciprocity is defined to cover only changes in manufacturing exports and imports, and not also changes in the trade of the nonmanufacturing good as would be consistent with our definition of reciprocity above, then tariff changes that conform to this adaptation of reciprocity keep the numbers of firms in each country unchanged and hence will be free of the firm delocation externality that countries would otherwise impose on one another with their trade policy choices; and for this reason reciprocity in Ossa's model behaves much like the terms-of-trade fixing property of reciprocity that we have emphasized above in the context of the terms-of-trade theory.

In particular, Ossa (2011) demonstrates that, beginning from Nash and under the restriction of nonnegative import tariffs, countries can gain from reciprocal tariff liberalization if they abide by this definition of reciprocity. And he shows that any point on the efficiency frontier is robust to renegotiation subject to reciprocity when reciprocity is defined in this way. More broadly, Ossa's findings point to the possibility that the principle of reciprocity can constitute a sensible design feature of trade agreements in environments that extend beyond the terms-of-trade theory, and at the same time suggest that in the absence of a full set of trade policy instruments it may be important to adopt a flexible view about the precise definition of reciprocity.cc

We consider next how the answers to the two questions above must be modified in environments where offshoring is prevalent. To begin, starting from their Nash tariffs, can both countries gain from at least a small amount of reciprocal liberalization? As it turns out, in the presence of offshoring this is no longer assured.

To see why, we return to the model of Antràs and Staiger (2012a). Recall that in (53) we defined the international input price px*; and as the world price of the final good is fixed on world markets by assumption, px*represents the terms of trade between the home and foreign country in this model. Hence, reciprocal reductions in τ1H, τxHand τxFare defined by any set of tariff reductions that hold px*fixed. Next recall that Nash policies are indeed inefficient in this model. But as Antràs and Staiger show, the Nash inefficiencies take a particular and interesting form: both the volume of input trade (xˆN) and the local price of the final good in the home-country market (p1HN) are inefficiently low. From this vantage point, it can now be seen that, beginning from the Nash point, even small reciprocal tariff cuts can lead to losses in this setting for a simple reason: if the home country is an exporter of the final good, then a reduction in τ1Hraises p1Hand therefore moves this price in the direction of the efficient level; but if the home country is an importer of the final good, then a reduction in τ1Hreduces p1Hand moves this price further away from the efficient level. Whether the resulting first-order loss in joint surplus can be made up with additions to joint surplus associated with the other elements of the reciprocal liberalization depends on circumstances, but with respect to the impact on p1Hit is clear that reducing τ1H(reciprocally or otherwise) reduces joint surplus when the home country is an importer of the final good.

Hence, in the presence of offshoring our earlier answer to the first question above must at a minimum be qualified. And our earlier answer to the second question is overturned in this environment, for the same reason that it is overturned in the missing-instruments environment: as with missing instruments, in the presence of offshoring it is no longer generally the case that politically optimal policies are internationally efficient, because as Antràs and Staiger (2012a) show and as we have described earlier, there is a local price externality that may now persist at the political optimum. In fact, as we have discussed, Antràs and Staiger argue that in the presence of offshoring the political optimum will be inefficient unless government objectives correspond to national income maximization. And with the inefficiency of the political optimum in the presence of offshoring comes as well the conclusion that no point on the efficiency frontier is robust to renegotiation subject to reciprocity. We therefore conclude that the attractive features of reciprocity that hold in the context of the terms-of-trade theory and that we have emphasized above may be substantially diminished in environments where offshoring is present.

Finally, notice that at one level all the models we have reviewed thus far share a common perspective on the purpose of a trade agreement: the trade agreement exists to address an international externality that is associated with unilateral policy choices. As we have described, regarding their stance on the question of purpose, where these models sometimes differ is in the form that the externality takes. The “commitment theory” of trade agreements is in this respect quite different, in that the central role for an international externality is absent. According to the commitment theory governments value trade agreements as a way to tie their hands (make commitments) against their own lobbies and citizens.cd Does reciprocity look attractive when viewed from the perspective of the commitment theory?

To answer this question, we begin by observing that as an international externality plays no fundamental role in the commitment theory, it is natural that the theory would predict no international inefficiencies associated with the unilateral policy choices of governments. Rather, in the commitment theory the inefficiencies associated with unilateral policy setting are domestic in nature: they reflect domestic distortions that are created when governments have incentives to surprise domestic private actors (producers and/or consumers) with unexpected policy intervention, and lack the ability to precommit on their own not to engage in such behavior. In the context of trade policy the incentive to surprise is especially likely to be present, owing to the second-best nature of trade-policy intervention and the incentive governments have to bring such intervention closer to the first best through the element of surprise (see, for example, Staiger and Tabellini, 1989). The point can be seen clearly in the case where a government uses a tariff to address a domestic consumption distortion, say, a negative externality (that does not cross borders) associated with domestic consumption. As a tariff is a combination production subsidy and consumption tax, it is a second-best instrument for addressing the consumption distortion; and it is for this reason that the government would have an incentive to surprise, announcing a policy of nonintervention until domestic production decisions had been made, and then following through with a tariff once production decisions are sunk so as to discourage consumption. In equilibrium domestic producers would not likely be fooled, and as a result the government might be better off if it could commit not to use the tariff at all. And if it cannot manage this commitment on its own, a trade agreement may serve as a useful external commitment device.

It should be fairly clear from this discussion that where governments seek to make commitments to their own private sectors through trade agreements, the attractive features of reciprocity that we have emphasized above in the context of the terms-of-trade theory are likely to lose their luster. This can be most directly appreciated by considering the case of a truly small country that seeks trade agreements as a way to make commitments to its private sectors. For such a country, reciprocity has no bite whatsoever, because by definition the trade policy choices of this country cannot alter the terms of trade; and so, if this country were allowed to renegotiate out of its commitments subject to reciprocity, its commitments would become meaningless. Put slightly differently, according to the logic of the commitment theory, governments care about their own tariff commitments but have no particular reason to care about the tariff commitments of their trading partners, whether those commitments are reciprocal or not. The one potential caveat to this statement arises in regard to how the commitments will be enforced. As trade agreements must generally be self-enforcing, and as commitments have no value if they are not enforced, it is possible that some form of reciprocity would be desired even by governments who saw the trade agreement as valuable to themselves only in so far as the agreement helped them to make commitments to their own private sectors. The logic is that only with reciprocity in some form would these governments (and their private sectors) expect that the commitments they undertook in a trade agreement might actually be enforced. We see this logic as a possible route along which reciprocity might be shown to play an important role in the commitment theory of trade agreements, and in that light as an interesting avenue for further research.

What is the movement of goods services and capital across national borders?

What are the four aspects of globalization?

Which of the following is not a characteristic of globalization?

What are the two major concerns of the European Union quizlet?