WPS5404 Policy Research Working Paper 5404 Novel Indicators of the Trade and Welfare Effects of Agricultural Distortions in OECD Countries Kym Anderson Johanna Croser The World Bank Development Research Group Agriculture and Rural Development Team August 2010 Novel Indicators of the Trade and Welfare Effects of Agricultural Distortions in OECD Countries Kym Anderson University of Adelaide kym.anderson@adelaide.edu.au and Johanna Croser University of Adelaide johanna.croser@adelaide.edu.au Keywords: Distorted incentives, agricultural price and trade policies, trade restrictiveness index JEL codes: F13, F14, Q17, Q18 Author contact: Kym Anderson School of Economics University of Adelaide Adelaide SA 5005 Australia Phone +61 8303 4712 Fax +61 8223 1460 kym.anderson@adelaide.edu.au This is a product of a World Bank research project on Distortions to Agricultural Incentives (see www.worldbank.org/agdistortions). The authors are grateful for helpful comments from referees, and for invaluable assistance with data compilation by Esteban Jara, Marianne Kurzweil, Signe Nelgen and Ernesto Valenzuela. Funding from World Bank Trust Funds provided by the governments of the Netherlands (BNPP) and the United Kingdom DfID), and from the Australian Research Council, is gratefully acknowledged. Views expressed are the authors' alone and not necessarily those of the World Bank or its Executive Directors. 2 Novel Indicators of the Trade and Welfare Effects of Agricultural Distortions in OECD Countries Empirical indicators of farm support by governments and their effects on consumer prices, called Producer and Consumer Support Estimates (PSEs and CSEs), have been estimated in a consistent way since 1986 by the Secretariat of the OECD (2009) for its 30 member countries. The indicators provide policy transparency, contribute to a better understanding of the various dimensions of agricultural support measures in high-income countries, and have been used extensively as inputs into economic models of agricultural markets. The OECD (2006) has also released PSEs for Brazil, China and South Africa, as well as for several East European countries; and it will soon add them for Chile. A recent global World Bank study (Anderson 2009) complements and extends the OECD's efforts by providing similar estimates for a longer time period (back to 1955) and for individual member countries of the European Union. It also has comparable estimates for 45 other countries at different stages of economic development and includes a time series of rates of assistance to producers of non- agricultural goods, to compare with agricultural distortion estimates. The OECD and World Bank measures for each product are aggregated using the value of production and consumption as weights to obtain an annual average PSE and CSE for each country. That traditional aggregation method provides a reasonable indicator of the average price distortion across that country's product set, but it is not necessarily a good indicator of the distortion to the volume of trade in farm products because that depends also on the responsiveness of domestic supply and demand to price changes (that is, price elasticities), and on whether there are any negative PSEs that are offsetting positive ones in the aggregating process. It is an even poorer indicator of the national welfare cost of that country's farm price and trade policies, because for each product that cost is related to the square of the rate of price distortion and so the total cost depends on the extent of dispersion in product PSEs and CSEs. 3 Certainly one can use the OECD or World Bank price distortions as inputs into national partial or general equilibrium models to estimate the trade- and welfare- reducing effects of a country's agricultural policies. However, such models are computationally intensive, and the results can be contentious if there is no consensus on what model specification and parameters such as elasticities to use. Even more problematic is that typically they are calibrated only for a particular past year and so are not able to provide a time series of estimated economic effects. An alternative is to use the raw data in the OECD and World Bank studies to calculate indexes of the trade- and welfare-reducing effects of policies. Anderson and Neary (2005) specify a simple, elegant and theoretically meaningful methodology to provide such measures as a supplement to aggregate PSEs and CSEs. The goal of this paper is to demonstrate how the Anderson-Neary methodology can be applied using no more information than that assembled already to generate price distortion estimates for OECD member countries. The method may have been ignored to date because it was traditionally thought that price elasticity estimates were necessary to estimate such indices. However, it has recently been shown by Lloyd, Croser and Anderson (2010) that by assuming domestic price elasticities of supply are equal across commodities within a country, and likewise for price elasticities of demand, the index number formulae simplify to a share-weighted function using shares of production and consumption as weights. The resulting measures thereby can be generated as supplements to the current policy monitoring indicators generated by the OECD Secretariat without having to tackle the contentious questions associated with the size of price elasticities (such as whether they refer to the short or long run) and without having to continually update a sector or economy- wide model. Drawing on the Anderson and Neary framework, we estimate two indexes which go by the precise descriptors of a trade reduction index (TRI) and a welfare reduction index (WRI). The TRI and WRI are each computed from sub-indices of the production and consumption sides of the market (the Producer and Consumer Distortion Indexes, PDI and CDI), which are derived from nominal rate of assistance (NRA) and consumer tax equivalent (CTE) estimates for individual products, 4 respectively, from the World Bank's database.1 NRAs to producers and CTEs to consumers differ whenever there are domestic subsidies or taxes on production or consumption in addition to border measures. Thus the indexes capture in a single scalar number the aggregate trade- or welfare-reducing effects of all policies directly affecting consumer and producer prices of farm products from all measures in place. Non-product-specific distortions are not captured in the indices, which by construction aggregate only product-specific data. However, we attempt to gauge the importance of this limitation in the final section of the paper. The present paper is aimed at encouraging not only the OECD to add these indexes to their current set of indicators calculated each year, but also developing country governments or policy think-tanks to generate them so as to be able to monitor each year the trade and welfare effects of their national policies. A new FAO/OECD project, funded by the Bill and Melinda Gates Foundation and getting under way in 2010, aims to estimate agricultural policy indicators for a large sample of poor African countries over the next few years. Since many of those countries do not have a sector or economy-wide model of their economy, the two indicators outlined in this paper could provide at least a partial equilibrium indication of the effect of national policies in reducing agricultural trade and national economic welfare. They could then be compared with those provided in the present paper for high-income countries. The paper begins with a presentation of the methodology for computing partial-equilibrium trade and welfare reduction indexes. It then outlines the data in the World Bank's database, which are used for computing the indices. Next, the index results are presented and discussed, following which is a section addressing several caveats. The paper concludes with lessons learned and areas for further research. Methodology There is a growing literature that identifies ways to measure the trade- and welfare- reducing effects of international trade policy in scalar index numbers. This literature serves a key purpose: it overcomes aggregation problems (across different 1 NRAs and CTEs are similar to PSEs and CSEs, except they are expressed as a percentage of the undistorted price whereas PSEs and CSEs are expressed as a percentage of the distorted price (and the CSE has the opposite sign to the CTE). 5 intervention measures and across industries) by using a theoretically sound aggregation procedure to answer precise questions regarding the trade or welfare reductions imposed by each country's agricultural or trade policies. The goal of the literature is to generate a single indicator that captures the overall trade or welfare effect of an individual country's regime of price distortions in place at any time, and to trace its path over time and make cross-country comparisons. The pioneering work in the literature is by Anderson and Neary (summarized in their 2005 book). Feenstra (1995) simplified the methodology to a partial- equilibrium framework. These two authors define a Trade Restrictiveness Index as the ad valorem trade tax rate which, if applied uniformly across all tradable agricultural commodities in a country, would generate the same reduction in welfare as the actual cross-product structure of distortions. They also define a Mercantilist Trade Restrictiveness Index (MTRI) as the ad valorem trade tax rate which, if applied uniformly across all tradable agricultural commodities in a country, would generate the same reduction in international trade as the actual cross-product structure of distortions. In recent years, several empirical papers have provided various series of partial-equilibrium estimates of scalar index numbers for individual countries. Irwin (2010) uses detailed tariff data to calculate the Trade Restrictiveness Index for the United States in 1859 and annually from 1867 to 1961. Kee, Nicita and Olarreaga (2009) estimate a series of indices for trade policies of 78 developing and developed countries for a single point in time (mid-2000s). Lloyd, Croser and Anderson (2010) modify the Anderson/Neary TRI and MTRI methodology to make it more applicable to agricultural policies, and show how it can be greatly simplified if certain assumptions about elasticities are adopted. Croser and Anderson (2010) build on that to develop a methodology for computing scalar index measures for individual policy instruments, which can be compared across instruments to see the relative contributions of different policy instruments to overall reductions in trade and welfare. In addition to being useful to summarize policy in an individual country, the Anderson-Neary scalar index measures has been adapted to measure the trade- and welfare-reducing effects of policy in a regional or global commodity market (Croser, Lloyd and Anderson 2010). In this paper we utilize the methodology in those latter three studies to generate a series of indicators of the trade- and welfare-reducing effects of agricultural policies in OECD countries over the past half century. 6 The remainder of the methodology section outlines the method for constructing three types of indexes: the Anderson-Neary type indexes for individual countries; instrument level indexes for individual countries to gauge the importance of different policy measures in the overall degree of agricultural policy distortions of OECD countries; and commodity market indexes for that group of countries. Country level trade and welfare reduction indexes To capture distortions imposed by each country's border and domestic policies on its economic welfare and its trade volume, we adopt the methodology from Lloyd, Croser and Anderson (2010). These authors define a Welfare Reduction Index (WRI) and a Trade Reduction Index (TRI), each of which can be estimated by considering separately the distortions to the producer and consumer sides of the economy (which can differ when there are domestic measures in place in addition to or instead of trade measures). As their names suggest, the two indexes respectively provide a single empirical indicator of the (partial equilibrium) welfare- or trade-reducing effects of distortions to consumer and producer prices of farm products from all agricultural and food policy measures in place. The Lloyd, Croser and Anderson (2010) methodology requires data on the production and consumption sides of the economy separately. Since PSE and CSE information is available from the OECD on an annual basis, this methodology is well suited to focusing on the trade and welfare effects of agricultural and trade policy in OECD member countries. Indeed it provides something closer than the PSE or CSE to what a sector or economy-wide computable general equilibrium model can provide in the way of estimates of the trade and welfare (and other) effects of price distortions, while having the advantage of providing an annual time series. The derivation of the measures in Lloyd, Croser and Anderson (2010) for n import-competing sectors leads to the expressions in Box 1 for the TRI and WRI for the import-competing sector of a country. The import-competing TRI and WRI are constructed from appropriately weighted averages of the level of distortions of consumer and producer prices. The same weights are used to construct both indexes, but the TRI is a mean of order one measure, while the WRI is a mean of order two. Because the WRI is a mean of order two, it better reflects the welfare cost of diverse agricultural price-distorting policies than the PSE or CSE since it captures the disproportionately higher welfare costs of peak levels of assistance or taxation. The 7 WRI is positive regardless of whether the government's agricultural policy is favoring or hurting farmers. The TRI and WRI can be readily extended to accommodate distortions to exported and nontradable agricultural goods (Lloyd, Croser and Anderson 2010). Separate sub-indices for each sub-sector are computed, and aggregated using sector values of production and consumption at undistorted prices as weights. Distortions to exportable industries enter the TRI aggregations as negative values because a positive (negative) price distortion in an exporting industry has a trade expanding (reducing) effect, and thus should decrease (increase) the TRI. Distortions to nontradable industries are assumed to be zero in the TRI aggregation because a domestic price distortion in a nontradable sector is assumed to have neither a trade expanding nor trade reducing effect because of the presence of high trade costs.2 Elasticities of supply and demand are required to compute the TRI and WRI expressions in Box 1. However, if one is willing to assume that price elasticities of supply (demand) are equal across commodities within a sub-sector or sector of an economy, then the elasticities in the numerator and denominator of the index weights cancel. This powerful simplifying assumption gives an expression for the TRI or WRI which is simply an appropriately weighted aggregate of distortions on production and consumption sides of the market. It is found by aggregating the change in consumer (producer) prices across commodities and using as weights the sector share of each commodity's domestic value of consumption (production) at undistorted prices. That is, with this elasticity assumption, these indexes are attainable with the same information used to estimate the PSE and CSE (or NRA and CTE, which are similar except expressed as a proportion of the border price rather than the distorted domestic price). Yet they provide a better indication of the trade- or welfare-distorting effects of those producer and consumer price measures. A second assumption is made in the empirical part of the paper when aggregating across all OECD countries. It is to assume that the marginal responses of a country's supply and demand to a price change are the same in aggregate for the sector. More precisely, we assume (see Box 1) that a=b=0.5, where the weight a (or b) is proportional to the ratio of the marginal response of domestic demand (or 2 This is consistent with the partial equilibrium nature of the indexes being generated here. In a general equilibrium model there could be indirect trade effects via the impact of distortions to nontradables on factor markets. 8 supply) to a price change relative to the marginal response of imports to a price change. Other trade and welfare reduction indexes The country level TRI and WRI measures reported below aggregate the trade- and welfare-reducing effects of a wide range of policy measures. The variables si and ri in Box 1, as domestic-to-border price ratios, can theoretically encompass distortions provided by all trade tax/subsidy measures and quantitative restrictions on trade, plus domestic price support measures (positive or negative), plus direct interventions on inputs; and, where multiple exchange rates operate (as in numerous developing countries in the past), the measures can encompass an estimate of the import or export tax equivalents of those distortions. While it is desirable to have a country level indicator that encompasses all of these distortions, agricultural policy analysts are sometimes interested in the relative contribution of different policy instruments to reductions in trade or welfare. To provide this insight, it is possible to use the Anderson-Neary framework to construct indicators of policy distortions at the instrument level and compare indices across instruments.3 Croser and Anderson (2010) define an Instrument Welfare Reduction Index (IWRI) and an Instrument Trade Reduction Index (ITRI), which can be estimated by considering the distortion from a single policy instrument to the producer and consumer sides of the economy. They develop their methodology for four types of border distortions (import taxes and subsidies, and export taxes and subsidies) and for a series of domestic distortions in the form of production, consumption and input taxes and subsidies. One of the limitations of the ITRI and IWRI in the context of OECD countries is that, by construction, non-product-specific measures are not included in the estimates because such supports are not reported at the product level. However, non- product-specific measures are clearly important for the overall story of agricultural policy in some OECD countries (reported below), as a result of a move in recent 3 This issue is not one that has been explored in the existing literature because most of the Anderson- Neary type indexes are estimated for single policy instruments. Irwin (2010), for example, uses only import tariffs. Kee, Nicita and Olarreaga (2009) report two series of indices, one based on tariffs only and the other on tariffs plus non-tariff import barriers. While they may be the dominant instruments for non-farm products, the agricultural sectors of OECD countries have been subject also to numerous domestic and export subsidies; and, in developing countries, agricultural production and export taxes also have been used. 9 decades to forms of support at least somewhat decoupled from production. Notwithstanding this limitation, below we estimate the trade- and welfare-reducing effects of individual policy instruments. We assume that border measures are applied first, and this may be supplemented by additional domestic distortions. This allocation assumption provides an upper-bound on welfare losses from border measures and a lower bound on welfare losses from domestic measures. An attempt is made in the empirical section below to gauge the potential importance of non-product-specific support measures which are excluded from the formal ITRI and IWRI measures. We also report commodity level TRI and WRI indexes below, which give the aggregate trade- and welfare-reducing effects of OECD member country policies to individual commodity markets. These indices are computed using a methodology similar to that in Box 1, but where distortions are summed across countries for an individual commodity, instead of across commodities for an individual country. Croser, Lloyd and Anderson (2009) provide a detailed exposition of the methodology as it applies to individual commodity markets globally. Below we provide them for the subset of countries that are OECD members. Data This study makes use of data from the World Bank's Distortions to Agricultural Incentives database (Anderson and Valenzuela 2008). For high-income countries that database drew on the OECD's PSE and CSE series (OECD 2009) for the period since 1986, but extended the time series back to 1955 for many countries. In the case of the European Union, whose membership expanded several times over the past half century, the World Bank study provides distortion estimates by country on the assumption that the estimated EU-wide PSE and CSE for each product applied in each member country (see Josling 2009). Differences across EU countries in the overall sector distortion indicators are thus due to differing commodity shares in sector production and consumption. We focus on a subset of OECD and other countries in the World Bank database (hereafter called the focus countries): 15 Western European countries, all of 10 which are OECD member countries;4 13 of Europe's transition economies, five of which are OECD member countries;5 and six other high-income OECD member countries: Australia, Canada, Japan, New Zealand, the Republic of Korea, and the United States. The OECD member countries that are not included in the focus countries sample are Belgium, Greece and Luxembourg (for which NRA estimates are not available and Mexico (a recent and much poorer member). The database contains annual estimates of nominal rates of assistance (NRAs, positive or negative) and consumer tax equivalents (CTEs) for key farm products. The NRA and CTE estimates in the database are at the commodity level and cover a subset of 39 agricultural products in the OECD. These so-called covered products account for around three-quarters of total agricultural production over the period studied. The database identifies the extent to which each commodity in each country each year is import or export dependent or a nontradable (which may change over time). For the 34 focus countries, the database contains around 16,000 consistent estimates of annual NRAs to the agricultural sector and the same number of CTEs between 1955 and 2007. Since not all countries have estimates for the 1950s, we report estimates below starting from 1960. The range of policy measures incorporated in the NRA estimates in the database is wide. By calculating domestic-to-border price ratios, the estimates include assistance provided by all tariff and nontariff trade measures at each country's border, plus any domestic price support measures (positive or negative), plus an adjustment for the output-price equivalent of direct interventions on inputs. Where and when multiple exchange rates operated, estimates of the import and export tax equivalents of that distortion are included as well. The range of measures included in the CTE estimates includes both domestic consumer taxes and subsidies and trade and exchange rate policies, all of which drive a wedge between the price that consumers pay for each commodity and the international price at the border. Analytical narratives of agricultural policies for the last five decades in the 34 OECD countries are provided in Anderson (2009). This book reports on the data in the Distortions to Agricultural Incentives database, and contains case studies for specific regional groupings. The book reports on measures such as unweighted and 4 Austria, Denmark, Finland, France, Germany, Iceland, Ireland, Italy, Netherlands, Norway, Portugal, Spain, Sweden, Switzerland and the UK. 5 Bulgaria, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Russia, Slovakia, Slovenia, Turkey and Ukraine (OECD member countries in bold). 11 weighted mean NRAs, standard deviations of NRAs, weighted mean NRAs for exportable versus import-competing covered products, measures of the trade bias of the agricultural sectors' covered plus non-covered tradable products, and relative rates of assistance. Josling (2009) provides an analysis of agricultural and trade policy distortions in Western Europe over the past 50 years. The analysis covers 18 countries, using data that has been disaggregated in some instances from regional aggregates. The aggregate NRA and CTE results from this study are reported at the country level in Tables 1 and 2. (Appendix Table 1 lists the changing membership of the regional EU and EFTA blocs.) It confirms that Western European agricultural policy is characterized by high levels of assistance throughout the postwar period, albeit with declines for some countries since the mid-1980s. The latter is largely due to some re-instrumentation of agricultural policy away from import protection for specific commodities toward direct payments that are supported for socially responsible farming. Anderson and Swinnen (2009) summarize agricultural policy in 18 of Europe's transition economies, drawing on their more-detailed book (Anderson and Swinnen 2008). Despite the heterogeneity of reform experiences, they note some overall patterns. In the early 1990s, when reliable data for these regions are first available, support to agriculture is at reasonably low levels. This is because many trade and price distortions were removed throughout the region at the start of the reform period in the early 1990s. Since that time, changes in agricultural policy have tended to be characterized overall by stop-go phases, and sometimes reversals of previous reforms. In 2000­07, NRAs were on average higher than they were in the decade of the 1990s. Honma and Hayami (2009) provide a study of agricultural policy in Northeast Asia over the past 50 years, illustrating the dramatic growth that can occur in distortions to agricultural incentives as real incomes grow. Distortions in these two countries are currently at high levels, driven by border protection for import-competing food products. North America's and Oceania's lower levels of agricultural policy distortion contrast with those of Europe and Northeast Asia. In the US and Canada, real spending on agricultural support has not diminished greatly over time (Gardner 2009). By contrast, in Australia and New Zealand, there was a rapid dismantling of 12 agricultural policy support from the 1980s, which has resulted in Oceania having the lowest levels of distortion among OECD countries (Anderson, Lattimore, Lloyd and MacLaren 2009). The country level aggregate measures in Tables 1 and 2 hide the degree of variation in commodity NRA and CTE estimates within countries. The case studies in Anderson (2009) report standard deviations around weighted mean NRAs for covered products in each country, showing that variation to be significant and not declining. One indication of the extent of variation between groups of products is provided by a comparison of the average NRA for import-competing and exportable product groups. The extent of this variation is shown in aggregate for EU countries and the 34 focus countries in Figure 1. There is a significant gap between the average NRA for import-competing and exportable products over the period shown, which reflects the extent of antitrade bias that has also persisted through time except for the most-recent period when international agricultural prices were rising. Notwithstanding the valuable contribution of the measures reported in the case studies in Anderson (2009), sector averages of NRAs can be misleading as indicators of the aggregate extent of price distortion within the sector as it affects trade and welfare. They can also be misleading when compared across countries which have varying degrees of dispersion in their NRAs for farm products. Hence the need for supplementary TRI and WRI series for the additional insights these measures can provide. TRI and WRI Estimates Table 3 reports the TRI estimates for all covered farm products from 1960 to 2007 for all 34 focus countries and six regional groupings. For all of the regional aggregations except Oceania, agricultural policy overall was trade-reducing, with Northeast Asia and Western Europe experiencing the largest reductions in trade. The regional aggregations hide some of the country level variation in agricultural policy, however, and there were even some decades in which policies were trade expanding in some countries, for example Finland, Sweden and several transition economies in addition to Australia and New Zealand. The TRI time series for the focus countries and the EU group are shown against the NRA time series in Figure 2. The most striking observation for these 13 groupings is the close correlation between the TRI and NRA series. This result is driven by the dominance of the import-competing sector in each of these two aggregations. The close correlation between the two series need not always result, however. Oceania provides a counter example, where the TRI has the opposite sign to the NRA aggregates, indicating that trade policy overall in Oceania was trade expanding despite positive NRA aggregates, because there was positive assistance to Australia's (and in some time periods New Zealand's) dominant export sub-sector. Another example of where the correlation between the TRI and NRA breaks down is in the time period 1980­84 to 1985-89. The NRA is increasing for 34 focus countries, from around 40 to 60 percent, while the TRI falls in this period by a similar amount. Agricultural policies in the focus countries were on aggregate becoming less trade restrictive in this period (even though the NRA is increasing) because assistance was increasing for exportable products, in the form of export subsidies. The WRI results reveal that over the period shown the aggregate NRA measure greatly understates the extent of welfare losses from agricultural and trade policies (Figure 2 and Table 4). Figure 2 shows that for EU countries the extent of understatement is greatest in the 1970s, and for the 34 focus countries the understatement is greatest in 1985­89. These large gaps coincide with world price- spikes. The 1985­89 period is when a downward price spike resulted in import- competing products being more distorted relative to export products, and conversely for the 1975­79 period. The fall in the WRI for EU countries is dramatic following the peak in the early 1980s, and more dramatic than the fall in the EU's aggregate NRA over the same time period (Figure 2). From the peak in 1980­84, there is a fall in both the weighted mean and the weighted variance of producer (consumer) distortions. Thus, the two elements of the WRI are falling, resulting in a steeper decline in the WRI than the NRA. This shows one of the benefits of generating a WRI: it provides a better sense of welfare improvements from policy reforms that reduce assistance to covered farm products. It should be noted, however, that from the mid-1980s, OECD members moved towards a re-instrumentation of agricultural policy, which is not fully reflected in the WRI and TRI estimates presented in Figure 2 (see next section). The individual country WRI results are presented in Table 4. They are necessarily always above the TRI and the average of the NRA and CTE measures, and are always positive because they are means of order two. There is considerable 14 variation in the extent of welfare reductions in policy over the period shown. In Western Europe, most countries have seen a decrease in their WRI in recent decades. For some countries this comes after a peak in the 1980s ­ such as in France, Ireland and Italy ­ whereas for other countries there has been more of a continual decline, as for example in the United Kingdom, Germany, Netherlands and Sweden. Norway, Switzerland and Iceland stand out among Western Europe countries for their exceptionally high WRIs, although these countries have experienced the steepest declines in recent decades. Canada's WRI series is notable for its large increase above the NRA aggregate in the 1980s (when there was a large increase in the dispersion of its NRAs around the weighted mean). The country-level WRI measures, which are derived using an overall measure of the distortion to producer and consumer prices in individual sectors of the 34 focus countries, masks the contribution of different policy instruments to welfare losses in each country. Figure 3 reports the decomposition of the overall country WRI by policy instrument for the 6 key regional groups. The decomposition is found by estimating WRI series for individual policy measures, and then apportioning the shares of these series to the overall WRI. In our 34 focus countries as a whole, border measures ­ which distort both producer and consumer prices ­ are by far the most significant of the distorting policy instruments. They account for upwards of 90 percent of the welfare losses in all 6 sub-regions over time, with the proportion being above 97 percent in most instances. The decomposition of border measures in Figure 3(a) shows that import tariffs are the dominant measure of distortion in terms of market price support in most regions. In the European Union and Northeast Asia, in particular, import taxes dominate border supports. In EFTA countries, import tariffs also dominate but these countries together also have significant export subsidies. Oceania has significant export subsidies in 1980­84, but they are reduced over time along with other reductions in policy distortions in those countries. Data are available only from 1992 for Eastern Europe's transition economies. In 2000­04, this sub-region has a range of distortionary policy instruments in use: import taxes dominate, but export taxes and subsidies are also present. The final perspective from which to consider the trade- and welfare-reducing effects of policies in our 34 focus countries is at the commodity market level, for individual commodities. Figure 4(a) shows that rice is the most distorted commodity 15 market across the 34 focus countries. This is followed by a group of vegetable products, which are heavily protected in Japan and Korea. The sugar, oilseed, milk, beef and cotton markets are the next most heavily distorted markets. The results for just the EU market indicate that sugar and livestock products are most heavily distorted in that region. Caveats and Sensitivity Analysis Some important caveats need to be mentioned, because the paper's two main indexes have been calculated with the help of a number of simplifying assumptions. One key assumption is that each country's own-price elasticity of supply (and also of demand) for a particular product is the same as that for every other product, and that cross-price elasticities are zero. It is not uncommon for modelers of the global market for particular farm products to adopt these assumptions, for want of reliable or agreed econometric estimates of those elasticities for each country (an early global example being Valdés and Zietz 1980). Anderson and Neary (2005, p. 293) observe that price elasticities are `not very influential' in affecting trade restrictiveness indices because elasticities appear in both the numerator and denominator of the indices (see Box 1). In the present case, too, this assumption is expected to have only a small effect on the results. Kee, Nicita and Olarreaga (2009) show that Anderson-Neary type indices can be decomposed into three elements: the weighted mean of distortions, the weighted variance of distortions, and the covariance between each distortion and its relevant elasticity scaled by the weighted average relevant elasticity. In empirical work, Kee, Nicita and Olarreaga (2009) note that the contribution of the covariance term to their estimated trade restrictiveness indexes is very small in practice. Irwin (2010) in his study for the United States similarly shows that the covariance is a very small factor relative to the average tariff and variance of the tariff. Notwithstanding those expectations, to gauge the potential importance of not allowing differential price responses, we re-compute our two country-level indexes using country- and commodity-specific own-price elasticity of supply and demand estimates available for 27 key farm products from widely cited sources (Roningen 2001; Tyers and Anderson 1992). A comparison in Table 5 of those results with the earlier estimates made with the simplifying elasticity assumption reveals some differences in the overall indications of distortions. The biggest divergences are for 16 Korea and Japan, where the average WRI across countries using the elasticity data is between 6 and 46 percentage points lower than estimates without elasticity data. It should be noted, however, that this is off a high base of WRI averages of over 100 percent in many instances. The Western European countries also have a fairly significant change in their TRI and WRI estimates. The elasticity values for this region reveal that livestock products tend to have a higher (absolute) elasticity of supply and demand, while grains and tropical crops have elasticities lower than the average.6 As such, including elasticity estimates results in livestock products in the EU having a higher weighting than grains and sugar. There is little divergence in the results with and without the simplifying elasticity assumption for North America and Oceania, which have relatively low TRI and WRI estimates. Despite the differences reported in Table 5, it is clear that in all cases, the index trends over time are much the same under either set of elasticity assumptions, and they give a better indication of the trade reduction and welfare losses from agricultural policies than standard weighted aggregates of NRAs and CTEs. Our other assumption -- that the aggregate marginal response of domestic demand to a price change is the same as the aggregate marginal response of domestic supply-- might also have an impact on the results. We re-compute our two indexes assuming that demand was instead twice, or half, as responsive as supply. Despite that wide range, the estimates were almost unchanged at the aggregate level across the six regional groups. This benign result is due to the empirical fact that the producer and consumer distortions are similar, reflecting the dominance of border measures in the policy instrument mix. A third caveat on the results for the TRI and WRI by policy instrument is the exclusion of non-product-specific (NPS) distortions in the estimates. In the Anderson and Valenzuela (2008) database, NPS assistance can be a significant component of overall agricultural sector distortions in some OECD countries. NPS is reported in three forms in the database: general NPS assistance, input subsidies that are not 6 Thus the size and ranking of the commodity indexes for the OECD country group, summarized in Figure 4, also would be affected somewhat by using differential elasticity estimates. Croser, Lloyd and Anderson (2010) examine this at the global level for eight major agricultural products and find that, if the elasticities found in Tyers and Anderson (1992) are used, there is little difference in the overall indications of distortions: the index averages using the elasticity estimates are 5 percentage points lower than the estimates using the simpler elasticity assumptions for one decade, but are between just 0 and 3 points lower for the other seven decade averages shown. Not surprisingly the differences are largest for the product with the most diverse NRAs, namely rice. In all cases, the global commodity index trends over time are much the same under either set of elasticity assumptions. 17 attributable at the product level, and decoupled payments. Recall that the ITRI (or IWRI) is defined as the ad valorem trade tax rate which, if applied uniformly across all tradable agricultural commodities in a country, would generate the same reduction in trade (or same economic welfare loss) as the actual cross-product structure of NRAs and CTEs for that country. A simple assumption to incorporate NPS measures is that all of the NPS distortions is enjoyed by producers and that they have no impact on consumer price distortions. This assumption allows us to provide, in Figure 3(b), an upper bound on their potential effect on the Producer Distortion Index (PDI) component of the ITRI or IWRI.7 Figure 3(b) shows the results of adding in this way all NPS assistance to the Producer Distortion Index. On the one hand, decoupled support and general NPS support ­ if equivalent to an increase in product prices for farmers ­ would make up almost one-third of distortions in EFTA and EU countries in 2000­04, and only slightly less in North America. On the other hand, if those forms of support were truly decoupled and had no impact on farmers' incentives, the PDI would be unaffected and hence the WRI would be as in Table 4. The potential importance of NPS for the WRI is thus somewhere in that range. The WRI and TRI series need to be interpreted in the light of the uncertainty associated with their omission of NPS measures. Conclusions This paper presents a case study of the application of new theory-based policy indicators to monitor the changing extent of policy interventions that reduce international trade and national economic welfare in OECD countries. It reports estimates of the indicators for each OECD country over the past half century as a way of illustrating the prospective use of this methodology as a supplement to the annually released PSE/CSE indicators of the OECD. The paper also shows that the methodology can be used to gain better insight into the trade and welfare reductions in individual commodity markets across OECD countries, and those reductions by individual policy instruments. 7 For example, if the IWRI of all border measures is 30 percent, and the country also gives farmers decoupled payment support of 20 percent, it is incorporated by assuming an overall country WRI of 50 percent. 18 In the past, trade and welfare reduction indicators have not been reported as part of the OECD regular monitoring activities. This may have been because it was thought that economic models and elasticity data would need to be agreed upon, which would raise technical and political problems. The measures we estimate in this paper are such that one can avoid the need to select a pair of price elasticity estimates for each product of each country. As such they could provide an attractive and politically uncontroversial supplement to the current policy monitoring indicators generated by the OECD, and by other multilateral institutions such as the FAO, UNCTAD, World Bank and the WTO. The importance of TRIs and WRIs will also be relevant for a new FAO/OECD project, funded by the Bill and Melinda Gates Foundation and getting under way in 2010, which aims to estimate agricultural policy indicators for a sample of African countries over the next few years. In African countries, different policy instruments operate such that the TRI could have a different sign in some years to the NRA aggregate (because of, for example, export taxes). Furthermore, if there is no economy-wide model for some of the African countries in the FAO/OECD project sample, the TRI and WRI can provide at least partial equilibrium indicators of the effect of national policies in reducing agricultural trade and national economic welfare. References Anderson, J.E. and J.P. Neary (2005), Measuring the Restrictiveness of International Trade Policy, Cambridge MA: MIT Press. Anderson, K and J.L. Croser, (2009) National and Global Agricultural Trade and Welfare Reduction Indexes, 1955 to 2007, World Bank, Washington DC, database available at www.worldbank.org/agdistortions Anderson, K. (2009), `Five Decades of Distortions to Agricultural Incentives', Ch. 1 in K. Anderson (ed.), Distortions to Agricultural Incentives: A Global Perspective, 1955-2007, London: Palgrave Macmillan and Washington DC: World Bank. 19 Anderson, K. and E. Valenzuela (2008), Global Estimates of Distortions to Agricultural Incentives, 1955 to 2007, database available at www.worldbank.org/agdistortions Anderson, K. and J. Swinnen (2008) (eds.), Distortions to Agricultural Incentives in Europe's Transition Economics, Washington DC: World Bank. Anderson, K. and J. Swinnen (2009), `Eastern Europe and Central Asia', Ch. 6 in K. Anderson (ed.), Distortions to Agricultural Incentives: A Global Perspective, 1955-2007, London: Palgrave Macmillan and Washington DC: World Bank. Anderson, K., R. Lattimore, P.J. Lloyd and D. MacLaren (2009), `Australia and New Zealand', Ch. 5 in Distortions to Agricultural Incentives: A Global Perspective, 1955-2007, London: Palgrave Macmillan and Washington DC: World Bank. Croser, J. and K. Anderson (2010), `Changing Contributions of Different Agricultural Policy Instruments to Global Reductions in Trade and Welfare', CEPR Discussion Paper 7748, London, March. Croser, J.L., P.J. Lloyd and K. Anderson (2010), `How Do Agricultural Policy Restrictions to Global Trade and Welfare Differ Across Commodities?' American Journal of Agricultural Economics 92(3): 698-712, April. Feenstra, R (1995). `Estimating the Effects of Trade Policy', in G. Grossman and K. Rogoff (eds.), Handbook of International Economics, Vol 3, Amsterdam: Elsevier. Gardner, B. (2009), `United States and Canada', Ch. 4 in K. Anderson (ed.), Distortions to Agricultural Incentives: A Global Perspective, 1955-2007, London: Palgrave Macmillan and Washington DC: World Bank. Honma, M. and Y. Hayami (2009), `Japan, Republic of Korea, and Tawian, China' Ch. 2 in K. Anderson (ed.), Distortions to Agricultural Incentives: A Global Perspective, 1955-2007, London: Palgrave Macmillan and Washington DC: World Bank. Irwin, D. (2010), `Trade Restrictiveness and Deadweight Losses from U.S. Tariffs, 1859­1961', American Economic Journal: Economic Policy 2: 119-45, August. Josling, T. (2009) `Western Europe', Ch. 3 in K. Anderson (ed.), Distortions to Agricultural Incentives: A Global Perspective, 1955-2007, London: Palgrave Macmillan and Washington DC: World Bank. 20 Kee, H.L., A. Nicita and M. Olerreaga (2009), `Estimating Trade Restrictiveness Indexes', Economic Journal 119(534): 172­199. Lloyd, P.J. (1974), `A More General Theory of Price Distortions in an Open Economy', Journal of International Economics 4(4): 365-86, November. Lloyd, P.J., J.L. Croser and K. Anderson (2010), `Global Distortions to Agricultural Markets: New Indicators of Trade and Welfare Impacts, 1960 to 2007', Review of Development Economics 14(2): 141-60, May. OECD (2006), Agricultural Policies, Markets and Trade in the Central and Eastern European Countries and the New Independent States: Monitoring and Outlook, Paris, OECD. OECD (2009), Producer and Consumer Support Estimates (online database accessed at www.oecd.org for 1986­2007 estimates; and OECD files for estimates using an earlier methodology for 1979­85. Roningen, V.O., `VORSIM version 5', www.vorsim.com. Serra, T., D. Zilberman, B. K. Goodwin and A. Featherstone (2006), `Effects of Decoupling on the Mean and Variability of Output', European Review of Agricultural Economics 33(3): 269­88. Tyers, R. and K. Anderson (1992), Disarray in World Food Markets: A Quantitative Assessment, Cambridge and New York: Cambridge University Press. Valdes, A. and J. Zeitz (1980), `Agricultural Protection in OECD Countries: its Cost to Less Developed Countries', Research Report 21, International Food Policy Research Institute, Washington DC, December. 21 Box 1: TRI and WRI expressions TRI WRI T {Ra Sb} , with W {R2 a S 2b}1/ 2 , with n n n 1/ 2 n 1/ 2 R rui and S si vi i R ' ri 2 u i and S ' s i2 vi i 1 i 1 i 1 i 1 where u i pi*2 dxi / dpiC / pi*2 dxi / dpiC i ( pi* xi ) / i ( pi* xi ) i i vi pi*2 dy i / dpiP / pi*2 dy i / dpiP i ( pi* y i ) / i ( pi* y i ) , i i a pi*2dx i / d p i / pi*2dmi / dpi , and b pi*2dyi / d p i / pi*2dmi / dpi . C P i i i i Variable definitions: T -- Trade Reduction Index; W -- Welfare Reduction Index; R -- weighted-average consumer price distortions; S -- weighted-average producer price distortions; R -- Consumer Distortion Index (CDI); S -- Producer Distortion Index (PDI); si -- the rate of distortion of the producer price in proportional terms; ri -- rate of distortion of the consumer price in proportional terms; ui -- weight for each commodity in R and R', which is proportional to the marginal response of domestic consumption to changes in international free-trade prices and can be written as a function of prices, demand quantities and domestic price elasticity (at the protected trade situation) of demand ( i ); vi -- weight for each commodity in S and S', which is proportional to the marginal response of domestic production to changes in international free-trade prices and can be written as a function of prices, supply quantities and domestic price elasticity (at the protected trade situation) of supply, ( i ); pi* -- border price; piP = pi*(1 + si ) -- distorted domestic price; p C = pi*(1 + ri ) -- distorted domestic consumer price; i xi xi ( pC ) -- quantity of good i demanded (as a function of own domestic price); yi yi ( pi ) -- quantity P i of good i supplied (as a function of own domestic price); a (b) -- weight of consumption (production) in the WRI or TRI, which is proportional to the ratio of the marginal response of domestic demand (supply) to a price change relative to the marginal response of imports to a price change. Source: Authors' compilation from Lloyd, Croser and Anderson (2010). 22 Table 1: Nominal rates of assistance, OECD countries, all covered products, 1960 to 2007 (percent) 1960-69 1970-79 1980-89 1990-99 2000-07 Western European Countries European Union 72 58 79 49 26 EFTA 62 56 111 178 144 Austria 53 21 40 66 33 Denmark 41 60 83 48 26 Finland 117 90 97 105 32 France 64 49 78 52 25 Germany 110 72 88 56 30 Iceland -- -- 277 219 137 Ireland 60 70 131 81 54 Italy 40 35 56 36 18 Netherlands 107 95 98 53 34 Norway -- -- 293 237 147 Portugal 11 22 30 29 19 Spain 16 -4 28 35 19 Sweden 134 90 92 75 32 Switzerland -- -- 296 258 143 UK 64 56 93 62 33 Europe's transition economies -- -- -- 9 18 Bulgaria -- -- -- -16 2 Czech Republic -- -- -- 17 21 Estonia -- -- -- 0 20 Hungary -- -- -- 16 21 Latvia -- -- -- 5 28 Lithuania -- -- -- 2 27 Poland -- -- -- 15 15 Romania -- -- -- 23 45 Russia -- -- -- 1 12 Slovakia -- -- -- 23 21 Slovenia -- -- -- 67 52 Turkey -- -- -- 20 24 Ukraine -- -- -- -13 -9 North America 6 6 15 10 10 Canada 8 11 26 17 13 US 6 6 14 9 10 Japan & Korea 62 87 135 156 143 Japan 73 94 133 148 132 Korea 10 61 145 192 189 Oceania 8 7 9 3 0 Australia 10 7 6 4 0 New Zealand 2 10 17 2 2 Source: Authors' calculations based on data in Anderson and Valenzuela (2008) 23 Table 2: Consumer tax equivalents, OECD Countries, all covered products, 1960 to 2007 (percent) 1960-69 1970-79 1980-89 1990-99 2000-07 Western European Countries European Union 71 57 68 38 24 EFTA 57 52 97 137 113 Austria 82 23 42 64 28 Denmark 41 68 74 43 24 Finland 128 92 123 124 31 France 64 52 69 39 24 Germany 101 67 70 38 24 Iceland -- -- 172 164 98 Ireland 42 84 120 64 35 Italy 43 36 52 32 20 Netherlands 103 89 97 52 30 Norway -- -- 57 115 101 Portugal 14 23 29 29 20 Spain 19 -2 20 27 18 Sweden 120 92 107 77 35 Switzerland -- -- 171 179 121 UK 55 52 83 49 33 Europe's transition economies -- -- -- 0 15 Bulgaria -- -- -- -15 5 Czech Republic -- -- -- 21 22 Estonia -- -- -- -1 15 Hungary -- -- -- 16 19 Latvia -- -- -- 15 32 Lithuania -- -- -- 1 25 Poland -- -- -- 3 22 Romania -- -- -- 5 34 Russia -- -- -- -12 19 Slovakia -- -- -- 14 17 Slovenia -- -- -- 53 36 Turkey -- -- -- 15 12 Ukraine -- -- -- -12 -1 North America 7 7 11 -1 -1 Canada 9 13 29 19 16 US 7 6 9 -4 -2 Japan & Korea 55 74 117 125 107 Japan 65 80 113 118 97 Korea 10 52 131 161 147 Oceania 12 11 10 6 2 Australia 17 11 8 7 2 New Zealand 3 10 15 3 2 Source: Authors' calculations based on data in Anderson and Valenzuela (2008) 24 Table 3: Trade reduction indexes, OECD countries, all covered products, 1960 to 2007 (percent) 1960-69 1970-79 1980-89 1990-99 2000-07 Western European Countries European Union 73 53 71 40 24 EFTA 40 27 24 27 57 Austria 67 22 2 15 30 Denmark -35 29 72 44 24 Finland 28 -8 -43 -51 31 France 66 46 70 41 23 Germany 105 65 77 45 26 Iceland -- -- 59 23 40 Ireland -8 51 123 72 44 Italy 47 33 50 28 18 Netherlands 104 91 97 52 32 Norway -- -- 60 175 120 Portugal 13 22 26 23 18 Spain 20 -1 23 27 17 Sweden 44 47 -11 -6 33 Switzerland -- -- 124 31 23 UK 59 50 86 54 33 Europe's transition economies -- -- -- 9 11 Bulgaria -- -- -- 10 8 Czech Republic -- -- -- -7 7 Estonia -- -- -- 11 4 Hungary -- -- -- -9 -17 Latvia -- -- -- 26 17 Lithuania -- -- -- 22 -5 Poland -- -- -- 10 -12 Romania -- -- -- 16 37 Russia -- -- -- -2 22 Slovakia -- -- -- 3 3 Slovenia -- -- -- -13 -20 Turkey -- -- -- 19 13 Ukraine -- -- -- 14 12 North America 4 3 8 5 4 Canada 7 10 22 17 13 US 3 2 7 4 3 Japan & Korea 58 81 126 140 118 Japan 69 87 123 133 112 Korea 10 56 138 177 144 Oceania -5 -4 -5 -3 0 Australia -9 -4 -5 -5 -1 New Zealand 2 -3 -6 2 1 Source: Authors' calculations based on data in Anderson and Valenzuela (2008) 25 Table 4: Welfare reduction indexes, OECD Countries, all covered products, 1960 to 2007 (percent) 1960-69 1970-79 1980-89 1990-99 2000-07 Western European Countries European Union 114 110 119 62 42 EFTA 125 111 145 181 148 Austria 92 41 60 83 47 Denmark 80 122 128 63 39 Finland 133 118 134 133 47 France 105 106 120 64 43 Germany 144 121 126 66 43 Iceland -- -- 274 238 167 Ireland 86 142 174 84 59 Italy 89 81 96 55 38 Netherlands 148 149 148 70 46 Norway -- -- 227 201 147 Portugal 26 44 50 49 37 Spain 44 33 59 51 35 Sweden 172 174 150 92 50 Switzerland -- -- 268 240 149 UK 144 127 132 72 50 Europe's transition economies -- -- -- 40 42 Bulgaria -- -- -- 27 25 Czech Republic -- -- -- 33 35 Estonia -- -- -- 27 28 Hungary -- -- -- 34 41 Latvia -- -- -- 50 52 Lithuania -- -- -- 53 53 Poland -- -- -- 28 34 Romania -- -- -- 40 60 Russia -- -- -- 39 34 Slovakia -- -- -- 31 33 Slovenia -- -- -- 69 57 Turkey -- -- -- 53 53 Ukraine -- -- -- 35 26 North America 16 14 35 23 23 Canada 15 32 83 46 38 US 17 12 30 20 22 Japan & Korea 77 119 190 221 192 Japan 84 130 198 225 190 Korea 44 77 153 202 203 Oceania 25 22 20 14 5 Australia 31 24 17 14 3 New Zealand 12 17 27 13 9 Source: Authors' calculations based on data in Anderson and Valenzuela (2008) 26 Table 5: Comparison of WRI and TRI estimates with and without simplifying elasticity assumption, sub-set of covered products, a 1960 to 2007 (percent) Using elasticity data for subset of With simplifying elasticity assumption Comparison (percentage point products for which data are available and for a sub-set of productsa difference between (1)-(5) and (6)-(10) (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) 1960- 1970- 1980- 1990- 2000- 1960- 1970- 1980- 1990- 2000- 1960- 1970- 1980- 1990- 2000- 69 79 89 99 07 69 79 89 99 07 69 79 89 99 07 Trade Reduction Indexes EU 68 51 65 34 22 75 55 74 40 24 7 3 10 6 2 EFTA 37 24 25 24 55 41 29 25 27 57 4 4 0 3 2 ECA na na na 5 7 na na na 8 11 - - - 4 4 NA 1 1 3 3 2 4 3 8 5 4 2 2 5 2 2 Japan & Korea 50 61 99 124 103 58 81 137 163 128 8 20 38 39 25 Oceania -2 -2 -4 -1 0 -5 -4 -6 -3 0 -3 -1 -2 -2 0 Welfare Reduction Indexes EU 103 102 109 57 40 116 113 123 62 42 13 11 14 6 2 EFTA 108 100 138 173 147 126 113 146 181 148 18 13 8 8 1 ECA na na na 39 45 na na na 41 43 - - - 2 -1 NA 13 11 27 19 20 16 14 35 23 23 4 3 9 4 4 Japan & Korea 70 101 162 201 171 77 119 201 246 205 6 18 39 46 34 Oceania 20 19 16 10 4 24 23 20 14 5 5 4 4 4 1 Sources: Authors' calculations based on data in Anderson and Valenzuela (2008) and elasticity estimates from Roningen (2001) and Tyers and Anderson (1992). a. The TRI and WRI estimates in these columns are for a sub-set of farm products for which we have elasticity data (so as to enable direct comparison of the results with and without the simplifying elasticity assumption). 27 Figure 1: Nominal rate of assistance, OECD countries, 1960 to 2007 (percent) (a) EU countries 120 100 80 60 40 20 0 1960-64 1965-69 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 2005-07 -20 All covered Exportables Import-competing (b) All OECD sample countries 100 90 80 70 60 50 40 30 20 10 0 1960-64 1965-69 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 2005-07 All covered Exportables Import-competing Sources: Authors' calculations based on data in Anderson and Valenzuela (2008). 28 Figure 2: Nominal rate of assistance, trade and welfare reduction indexes, OECD countries, 1960 to 2007 (percent) (a) EU countries 140 120 100 80 60 40 20 0 1960-64 1965-69 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 2005-07 NRA WRI TRI (b) All OECD sample countries 160 140 120 100 80 60 40 20 0 1960-64 1965-69 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 2005-07 NRA WRI TRI Sources: Authors' calculations based on data in Anderson and Valenzuela (2008). 29 Figure 3: Decomposition of indices by policy instrument, 1980-84 and 2000-04 (percent) (a) Decomposition of WRI, border measure only components 1980­84 2000­04 250 250 200 200 150 150 100 100 50 50 0 0 ANZ NA ECA EU EFTA Ja/Ko ANZ NA ECA EU EFTA Ja/Ko Import tax Export tax Import tax Export tax Export subsidy Import Subsidy Export subsidy Import Subsidy (b) Decomposition of PDI, border and domestic components 1980­84 2000­04 350 350 300 300 250 250 200 200 150 150 100 100 50 50 0 0 ANZ NA ECA EU EFTA Ja/Ko ANZ NA ECA EU EFTA Ja/Ko Border measures Border measures Production tax Production tax Production subsidy Production subsidy Input subsidy Input subsidy Decoupled payments Decoupled payments Nonproduct specific (NPS) support Nonproduct specific (NPS) support NPS to inputs NPS to inputs Source: Authors' calculations based on data in Anderson and Valenzuela (2008). 50 0 100 150 200 250 300 350 400 450 500 Coarse grains Other crops Banana Fruit & ... Olive Peas (a) 34 focus countries Orange Wine Rapeseed Tomato Wool Other grains Hazelnut Sorghum Tobacco Sunflower Strawberry Maize Wheat Apple Rye 30 198084 Egg (percent) Potato Sheepmeat Oat Figure 4: OECD commodity market welfare reduction indexes, 1980­84 and 2000­04 Soybean 200004 Mandarin Barley Pigmeat Cucumber Poultry Cotton Beef Grape Milk Oilseed Cabbage Garlic Spinach Sugar Pear Pepper Onion Rice 31 Figure 4 (continued): OECD commodity market welfare reduction indexes, 1980­84 and 2000­04 (percent) (b) Western European countries 350 300 250 200 150 100 50 0 o to an e t y at Ra er r y t t ef ed ilk e ce g ea ea Oa ga at tr rle aiz in Eg he Be ta w M Ri be se ul m W Su pm gm Ba lo M Po W Po pe y To nf So Pi ee Su 198084 200004 Sh Source: Authors' calculations based on data in Anderson and Valenzuela (2008). 32 Appendix Table 1: EU and EFTA members represented in the Agricultural Distortions database (a) European Union (EU) members representeda Year Countries 1956 France, Germany, Italy, Netherlands, 1973 plus UK, Ireland, Denmark 1986 plus Portugal, Spain 1995 plus Austria, Sweden, Finland (a) European Free Trade Association (EFTA) members represented Year Countries 1960 Austria, Denmark, Norway, Portugal, Sweden, Switzerland, UK 1970 Austria, Denmark, Norway, Portugal, Sweden, Switzerland, UK, Finland, Iceland 1973 Austria, Norway, Portugal, Sweden, Switzerland, Finland and Iceland 1986 Austria, Norway, Sweden, Switzerland, Finland and Iceland 1995 Norway, Switzerland and Iceland a Several of Europe's transition economies joined the EU in 2004 and 2007. These countries are not included in the EU aggregates provided in this paper, but instead are included as part of the Eastern Europe and Central Asia (ECA) aggregation. Also not included at Cyprus and Malta, which joined the EU in 2004. Source: Authors' aggregations. 33 Appendix Table 2: OECD commodity market trade reduction indexes, 44 covered farm products, 1960 to 2004 (percent) 1960-64 1965-69 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 Grains 22 24 13 21 25 64 63 38 35 Barley 36 31 3 -14 -1 37 32 10 4 Coarse grains -4 -4 -4 -4 -4 -2 0 0 0 Maize 3 6 3 10 4 7 12 7 7 Oat 15 9 -8 -3 -10 -2 -2 13 9 Other crops 0 0 0 0 0 0 0 0 0 Other grains na na na na na na 10 18 12 Rice 60 73 95 125 165 357 404 326 374 Rye na na na na na na 2 1 14 Sorghum 0 0 0 -1 -15 -3 6 5 7 Wheat 17 15 -3 0 12 31 28 5 7 Oilseeds 0 0 -1 2 6 15 11 5 1 Hazelnut na na na 17 57 47 40 31 4 Oilseed na na na 310 343 468 286 41 47 Rapeseed -4 -2 -1 -1 -1 31 16 0 0 Soybean 0 1 -1 2 5 6 5 2 0 Sunflower 0 -5 -12 -16 -31 41 28 16 11 Tropical Crops 29 61 2 35 50 63 43 47 46 Cotton 1 -44 -29 -6 -8 -4 2 6 -12 Sugar 102 217 17 81 109 164 99 105 111 Tobacco 45 48 68 45 63 11 -25 -37 11 Livestock 39 41 37 47 58 52 37 37 34 Beef 24 21 18 16 39 53 41 45 42 Egg -8 -4 -7 10 9 14 11 13 8 Milk 88 92 87 143 148 133 76 73 56 Pigmeat 27 37 29 26 34 11 7 15 15 Poultry 22 20 28 26 26 25 29 25 26 Sheepmeat 64 80 110 167 98 72 43 20 20 Wool 0 0 -6 -4 -7 -2 -4 0 0 Fruit &vegetables 11 5 7 18 13 10 8 12 11 Apple -6 8 28 44 43 21 19 14 18 Banana 0 0 0 0 5 1 0 0 0 Cabbage na na na na na 17 28 79 90 Cucumber na na na na na 57 17 30 43 Fruit & vegetables 0 0 0 0 0 0 0 0 0 Garlic na na na na na 250 289 213 123 Grape 7 10 -4 5 8 16 18 31 22 Mandarin na na na na na 21 45 47 32 34 1960-64 1965-69 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 Olive 0 0 0 0 0 0 0 0 0 Onion na na na na na 55 81 144 284 Orange 25 25 26 33 38 13 3 1 1 Pear na na na na na 35 24 64 157 Peas 0 0 0 0 0 0 0 0 0 Pepper na na na na na 175 245 146 197 Potato 24 19 16 48 27 9 8 4 0 Spinach na na na na na 89 138 237 134 Strawberry na na na na na 11 25 26 17 Tomato -4 20 21 21 19 8 -5 3 2 Wine 10 -3 -3 -4 -9 -2 -10 -4 -2 Source: Authors' calculations based on data in Anderson and Valenzuela (2008). 35 Appendix Table 3: OECD commodity market welfare reduction indexes, 44 covered farm products, 1960 to 2004 (percent) 1960-64 1965-69 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 Grains 39 48 38 47 48 103 96 61 55 Barley 52 49 35 41 32 98 88 45 33 Coarse grains 4 4 4 4 4 2 0 0 0 Maize 16 21 16 23 23 29 26 15 17 Oat 52 72 63 105 41 67 70 33 31 Other crops 0 0 0 0 0 0 0 0 0 Other grains na na na na na na 11 18 12 Rice 75 92 118 149 192 429 481 391 447 Rye na na na na na na 35 26 25 Sorghum 0 0 0 2 24 12 9 11 15 Wheat 35 45 25 21 31 63 57 28 23 Oilseeds 4 6 9 14 31 41 34 24 26 Hazelnut na na na 21 57 47 40 31 12 Oilseed na na na 354 378 472 352 113 103 Rapeseed 17 7 4 3 2 62 44 4 2 Soybean 4 6 9 14 31 32 29 28 31 Sunflower 10 9 16 24 43 69 49 23 16 Tropical Crops 49 133 46 52 69 99 74 77 90 Cotton 7 58 39 16 19 40 36 38 55 Sugar 161 288 46 100 126 192 123 128 139 Tobacco 46 48 97 52 63 22 36 45 15 Livestock 78 80 72 86 92 94 75 64 61 Beef 47 39 37 45 76 102 86 83 80 Egg 46 46 31 19 20 40 37 38 27 Milk 161 163 149 233 202 211 127 102 86 Pigmeat 50 77 63 57 67 35 35 32 34 Poultry 37 33 46 39 43 48 58 48 51 Sheepmeat 103 144 180 216 140 116 76 39 30 Wool 0 0 6 7 11 7 10 8 6 Fruit &vegetables 28 20 17 30 26 20 21 19 22 Apple 6 16 35 49 48 24 21 21 24 Banana 0 0 0 0 5 1 0 0 0 Cabbage na na na na na 21 34 93 116 Cucumber na na na na na 57 17 30 43 Fruit & vegetables 0 0 0 0 0 0 0 0 0 Garlic na na na na na 250 289 213 123 Grape 59 29 19 11 13 42 51 64 81 Mandarin na na na na na 21 45 47 32 36 1960-64 1965-69 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 Olive 0 0 0 0 0 0 0 0 0 Onion na na na na na 55 81 144 284 Orange 25 25 26 33 38 13 3 1 1 Pear na na na na na 35 24 64 157 Peas 0 0 0 0 0 0 0 0 0 Pepper na na na na na 175 245 146 197 Potato 80 79 35 74 45 17 17 13 28 Spinach na na na na na 89 138 237 134 Strawberry na na na na na 11 25 26 17 Tomato 16 25 28 26 23 19 8 9 6 Wine 18 4 4 4 10 11 10 4 2 Source: Authors' calculations based on data in Anderson and Valenzuela (2008). 37 Appendix Table 4: Elasticities of supply, 27 key covered farm products, OECD member countries and transition economies Australia Austria Bulgaria Canada Czech Rep Denmark Estonia Finland France Germany Hungary Iceland Barley 0.93 0.60 0.80 0.99 0.80 0.70 0.80 0.60 0.70 0.70 0.80 - Beef 0.70 0.57 0.30 0.50 0.30 0.55 0.30 0.57 0.55 0.55 0.30 0.57 Cotton 0.50 - - - - - - - - - - - Egg 0.60 0.75 0.35 0.55 0.35 0.75 0.35 0.75 0.75 0.75 0.35 0.75 Hazelnut - - - - - - - - - - - - Maize 0.80 0.65 0.30 0.48 0.30 - - - 0.60 0.60 0.30 - Milk 0.50 0.60 0.30 0.50 0.30 0.65 0.30 0.60 0.65 0.65 0.30 0.60 Oat 0.93 0.60 0.80 - 0.80 0.70 0.80 0.60 0.70 0.70 0.80 - Oilseeds 0.60 - - - - - 0.30 - - - - - Pigmeat 0.80 0.80 0.45 1.50 0.45 0.90 0.45 0.80 0.90 0.90 0.45 0.80 Potato 0.93 0.60 0.80 0.99 0.80 0.70 0.80 0.60 0.70 0.70 0.80 - Poultry 0.80 0.75 0.70 0.70 0.70 0.80 0.70 0.75 0.80 0.80 0.70 0.75 Rapeseed 0.60 0.30 0.30 0.55 0.30 0.75 - - 0.75 0.75 0.30 - Rice 0.60 - 0.30 - - - - - 0.35 - 0.30 - Rye - - - - - - 0.80 - - - - - Sheepmeat 0.70 0.80 0.35 - 0.35 0.70 0.35 0.80 0.70 0.70 0.35 0.80 Sorghum 0.93 - - - - - - - - - - - Soybean 0.50 - 0.45 0.60 0.45 - - - 0.40 0.40 0.45 - Sugar 0.50 0.45 0.20 0.50 0.20 0.15 - 0.45 0.15 0.15 0.20 - Sunflower 0.60 0.30 0.30 - 0.30 - - - 0.75 0.75 0.30 - Tobacco 0.50 - - - - - - - - - - - Wheat 0.90 0.80 0.25 0.60 0.25 0.50 0.25 0.80 0.50 0.50 0.25 - Wine - 0.20 0.20 - - - - - 0.20 0.20 0.20 - Wool 0.70 - - - - - - - - - - 0.80 38 New Ireland Italy Japan Korea Latvia Lithuania Netherlands Zealand Norway Poland Portugal Romania Barley 0.70 0.70 0.62 0.37 0.80 0.80 0.70 0.80 0.60 0.80 0.70 0.80 Beef 0.55 0.55 0.40 0.50 0.30 0.30 0.55 0.45 0.57 0.30 0.55 0.30 Cotton - - - - - - - - - - - - Egg 0.75 0.75 0.80 0.80 0.35 0.35 0.75 0.80 0.75 0.35 0.75 0.35 Hazelnut - - - - - - - - - - - - Maize - 0.60 - - - 0.30 0.60 0.90 - 0.30 0.60 0.30 Milk 0.65 0.65 0.40 0.80 0.30 0.30 0.65 0.60 0.60 0.30 0.65 0.30 Oat 0.70 0.70 - - 0.80 0.80 0.70 0.80 0.60 0.80 0.70 0.80 Oilseed - - - - 0.30 0.30 - - - 0.30 - - Pigmeat 0.90 0.90 0.88 0.70 0.45 0.45 0.90 0.80 0.80 0.45 0.90 0.45 Potato 0.70 0.70 - - 0.80 0.80 0.70 - - 0.80 0.70 0.80 Poultry 0.80 0.80 1.27 0.90 0.70 0.70 0.80 0.80 0.75 0.70 0.80 0.70 Rapeseed 0.75 0.75 - - - - 0.75 - - - - 0.30 Rice - 0.35 0.50 0.35 - - - - - - 0.35 0.30 Rye - - - - 0.80 0.80 - - - - - - Sheepmeat 0.70 0.70 - - 0.35 0.35 0.70 0.90 0.80 0.35 0.70 0.35 Sorghum - - - - - - - - - - - - Soybean - 0.40 0.65 0.36 - - - - - 0.45 - 0.45 Sugar 0.15 0.15 0.45 - 0.20 0.20 0.15 - - 0.20 0.15 0.20 Sunflower - 0.75 - - - - - - - 0.30 0.75 0.30 Tobacco - - - - - - - - - - - - Wheat 0.50 0.50 0.52 0.45 0.25 0.25 0.50 0.80 0.80 0.25 0.50 0.25 Wine - 0.20 - - - - - - - - 0.20 0.20 Wool - - - - - - - 0.90 0.80 - - - 39 Russia Slovakia Slovenia Spain Sweden Switzerland Turkey UK Ukraine US Barley 0.24 0.80 0.80 0.70 0.60 0.60 0.80 0.70 0.80 0.99 Beef 0.25 0.30 0.30 0.55 0.57 0.57 0.30 0.55 0.30 0.60 Cotton - - - - - - 0.15 - - 0.74 Egg 0.25 0.35 0.35 0.75 0.75 0.75 0.35 0.75 0.35 0.55 Hazelnut - - - - - - 0.20 - - - Maize 0.38 0.30 0.30 0.60 - 0.65 0.30 - 0.30 0.48 Milk 0.20 0.30 0.30 0.65 0.60 0.60 0.30 0.65 0.30 0.50 Oat 0.24 0.80 - 0.70 0.60 0.60 - 0.70 0.80 - Oilseed - - - - - 0.30 - - - - Pigmeat 0.40 0.45 0.45 0.90 0.80 0.80 - 0.90 0.45 1.00 Potato - 0.80 - 0.70 0.60 - 0.80 0.70 0.80 0.99 Poultry 0.50 0.70 0.70 0.80 0.75 0.75 0.70 0.80 0.70 0.65 Rapeseed - 0.30 - 0.75 0.30 - - 0.75 - - Rice - - - 0.35 - - 0.30 - - 0.40 Rye 0.24 0.80 - - - - - - 0.80 - Sheepmeat - 0.35 0.35 0.70 0.80 0.80 0.35 0.70 - 0.80 Sorghum - - - - - - - - - 0.99 Soybean - 0.45 - 0.40 - - - - - 0.60 Sugar 0.16 0.20 0.20 0.15 0.45 0.45 0.20 0.15 0.20 0.50 Sunflower 0.15 0.30 - 0.75 - - 0.30 - 0.30 - Tobacco - - - - - - 0.20 - - - Wheat 0.23 0.25 0.25 0.50 0.80 0.80 0.25 0.50 0.25 0.60 Wine - 0.20 - 0.20 - - - - - - Wool - - - - - - - - - 0.80 Sources: Authors' compilation from Roningen (2001) and Tyers and Anderson (1992, Appendix Tables A2 to A4). 40 Appendix Table 5: Elasticities of demand (absolute value), 27 key covered farm products, OECD member countries and transition economies Australia Austria Bulgaria Canada Czech Rep Denmark Estonia Finland France Germany Hungary Iceland Barley 0.64 0.77 0.68 1.07 0.68 0.91 0.68 0.77 0.91 0.91 0.68 - Beef 0.78 0.70 0.20 0.80 0.20 0.70 0.20 0.70 0.70 0.70 0.20 0.70 Cotton 0.20 - - - - - - - - - - - Egg 0.25 0.35 0.10 0.30 0.10 0.20 0.10 0.35 0.20 0.20 0.10 0.35 Hazelnut - - - - - - - - - - - - Maize 0.68 1.11 0.66 1.03 0.66 - - - 1.06 1.06 0.66 - Milk 0.16 0.15 0.14 0.15 0.14 0.16 0.14 0.15 0.16 0.16 0.14 0.15 Oat 0.64 0.77 0.68 - 0.68 0.91 0.68 0.77 0.91 0.91 0.68 - Oilseeds 0.38 - - - - - 0.72 - - - - - Pigmeat 1.02 0.60 0.50 0.86 0.50 0.80 0.50 0.60 0.80 0.80 0.50 0.60 Potato 0.64 0.77 0.68 1.07 0.68 0.91 0.68 0.77 0.91 0.91 0.68 - Poultry 0.80 0.65 0.25 0.67 0.25 0.90 0.25 0.65 0.90 0.90 0.25 0.65 Rapeseed 0.38 0.27 0.72 0.45 0.72 0.35 - - 0.35 0.35 0.72 - Rice 0.45 - 0.15 - - - - - 0.50 - 0.15 - Rye - - - - - - 0.68 - - - - - Sheepmeat 1.20 0.47 0.28 - 0.28 0.90 0.28 0.47 0.90 0.90 0.28 0.47 Sorghum 0.64 - - - - - - - - - - - Soybean 0.25 - 0.13 0.26 0.13 - - - 0.16 0.16 0.13 - Sugar 0.25 0.29 0.30 0.24 0.30 0.50 - 0.29 0.50 0.50 0.30 - Sunflower 0.38 0.27 0.72 - 0.72 - - - 0.35 0.35 0.72 - Tobacco 0.50 - - - - - - - - - - - Wheat 0.39 0.53 0.37 0.57 0.37 0.57 0.37 0.53 0.57 0.57 0.37 - Wine - 0.50 0.50 - - - - - 0.50 0.50 0.50 - Wool 1.20 - - - - - - - - - - 0.47 41 New Ireland Italy Japan Korea Latvia Lithuania Netherlands Zealand Norway Poland Portugal Romania Barley 0.91 0.91 1.32 0.74 0.68 0.68 0.91 0.64 0.77 0.68 0.91 0.68 Beef 0.70 0.70 1.00 0.80 0.20 0.20 0.70 0.60 0.70 0.20 0.70 0.20 Cotton - - - - - - - - - - - - Egg 0.20 0.20 0.30 0.20 0.10 0.10 0.20 0.60 0.35 0.10 0.20 0.10 Hazelnut - - - - - - - - - - - - Maize - 1.06 - - - 0.66 1.06 0.89 - 0.66 1.06 0.66 Milk 0.16 0.16 0.19 0.80 0.14 0.14 0.16 0.01 0.15 0.14 0.16 0.14 Oat 0.91 0.91 - - 0.68 0.68 0.91 0.64 0.77 0.68 0.91 0.68 Oilseed - - - - 0.72 0.72 - - - 0.72 - - Pigmeat 0.80 0.80 0.95 0.90 0.50 0.50 0.80 0.55 0.60 0.50 0.80 0.50 Potato 0.91 0.91 - - 0.68 0.68 0.91 - - 0.68 0.91 0.68 Poultry 0.90 0.90 1.10 0.70 0.25 0.25 0.90 0.60 0.65 0.25 0.90 0.25 Rapeseed 0.35 0.35 - - - - 0.35 - - - - 0.72 Rice - 0.50 0.25 0.20 - - - - - - 0.50 0.15 Rye - - - - 0.68 0.68 - - - - - - Sheepmeat 0.90 0.90 - - 0.28 0.28 0.90 0.60 0.47 0.28 0.90 0.28 Sorghum - - - - - - - - - - - - Soybean - 0.16 0.14 0.12 - - - - - 0.13 - 0.13 Sugar 0.50 0.50 0.54 - 0.30 0.30 0.50 - - 0.30 0.50 0.30 Sunflower - 0.35 - - - - - - - 0.72 0.35 0.72 Tobacco - - - - - - - - - - - - Wheat 0.57 0.57 0.39 0.61 0.37 0.37 0.57 0.31 0.53 0.37 0.57 0.37 Wine - 0.50 - - - - - - - - 0.50 0.50 Wool - - - - - - - 0.60 0.47 - - - 42 Russia Slovakia Slovenia Spain Sweden Switzerland Turkey UK Ukraine US Barley 0.38 0.68 0.68 0.91 0.77 0.77 0.68 0.91 0.68 1.38 Beef 0.19 0.20 0.20 0.70 0.70 0.70 0.20 0.70 0.20 0.70 Cotton - - - - - - 0.15 - - 0.20 Egg 0.15 0.10 0.10 0.20 0.35 0.35 0.10 0.20 0.10 0.35 Hazelnut - - - - - - 0.50 - - - Maize 0.54 0.66 0.66 1.06 - 1.11 0.66 - 0.66 0.80 Milk 0.15 0.14 0.14 0.16 0.15 0.15 0.14 0.16 0.14 0.16 Oat 0.38 0.68 - 0.91 0.77 0.77 - 0.91 0.68 - Oilseed - - - - - 0.27 - - - - Pigmeat 0.18 0.50 0.50 0.80 0.60 0.60 - 0.80 0.50 0.86 Potato - 0.68 - 0.91 0.77 - 0.68 0.91 0.68 1.38 Poultry 0.25 0.25 0.25 0.90 0.65 0.65 0.25 0.90 0.25 0.56 Rapeseed - 0.72 - 0.35 0.27 - - 0.35 - - Rice - - - 0.50 - - 0.15 - - 0.25 Rye 0.38 0.68 - - - - - - 0.68 - Sheepmeat - 0.28 0.28 0.90 0.47 0.47 0.28 0.90 - 0.70 Sorghum - - - - - - - - - 1.38 Soybean - 0.13 - 0.16 - - - - - 0.30 Sugar 0.15 0.30 0.30 0.50 0.29 0.29 0.30 0.50 0.30 0.24 Sunflower 0.37 0.72 - 0.35 - - 0.72 - 0.72 - Tobacco - - - - - - 0.50 - - - Wheat 0.29 0.37 0.37 0.57 0.53 0.53 0.37 0.57 0.37 0.49 Wine - 0.50 - 0.50 - - - - - - Wool - - - - - - - - - 0.70 Sources: Authors' compilation from Roningen (2001) and Tyers and Anderson (1992, Appendix Tables A2 to A4).