Barriers to trade between the EU and Japan

Bertin MARTENS
EU-Japan trade relations have been very active for many years. There has been a dialogue between Japan and the EU on regulatory issues in this regard. Regulatory issues, or non-tariff measures (NTMs), were focused on rather than tariffs in this report because tariffs are generally very low between Japan and the EU. What was missing in the regulatory dialogue was an economic discussion of the costs of regulations, thus prompting this study.

One of the first things that was found in the study was that the intensity of the trade relationship between the EU and Japan has been declining over the last 10 years. Japan's share of the EU's total external trade in terms of both imports and exports has declined. The same is true of the EU's share of Japan's external trade, though the decline is not as strong. While increasing trade with emerging markets for both the EU and Japan is a possible explanation for this, another explanation is that regulatory issues and barriers have become important.

From a European perspective, Japan is a difficult market for EU exports. The ratio of exports divided by destination countrys' GDP shows that the average for the EU is around 2.6%, but for Japan, it is 1.7%, which can be seen as a kind of trade gap. Japan is not only a difficult market for EU exports, but also for other economies. Japan's import share as a percentage of domestic demand, at 6%, is among the lowest in the OECD, showing that import penetration in Japan is very low.

Japan's exports are very much concentrated in a few sectors, as demonstrated by the fact that the four most important sectors of Japanese exports to the EU (cars, electronics, machinery, and office equipment) account for 80% of the total. The historic explanation for this is that Japan's exports are still very much concentrated on sectors that METI was very active in promoting around 30 years ago. EU exports to Japan are more broadly distributed over a wider range of sectors. Car exports to Japan is the most important sector, something that is easily forgotten.

Tariffs are relatively low with a weighted average of 3.8% as the EU and Japan share most favored nation tariffs with each other. However, some have begun to call this "least favored nation" tariffs, since many countries no longer adhere to most favored nation tariffs. In fact, 75% of EU imports enter the EU duty-free due to preferences that have been awarded through bilateral trade agreements. Tariff peaks exist for the EU on products such as cars, trucks, textiles, and agricultural products, and Japan also has peak tariffs in agricultural products.

The study concentrated on the costs of NTMs in Japan. The question was how to measure the importance of these measures and their impact on trade. A survey was launched targeting EU-based companies operating in Japan to collect empirical estimates of the economic costs attached to regulatory measures. The list started with 100 NTMs, most of which applied to the pharmaceutical, food, and office equipment industries, among other sectors. The measures can be classified into three categories. The first category includes measures that can be addressed by trade policy instruments. The second category includes measures that require trade policy instruments as well as domestic reforms to be addressed. The third category includes measures that can only be addressed by domestic reforms.

The study shows that there are only a limited number of measures that can be addressed through trade agreements and some of them need to be complemented by domestic reforms, indicating that trade agreements alone cannot alone solve all trade issues. Also, conformity assessment problems dominated in the list of measures, taking up two thirds of all measures. Similarly for the service sector, most measures require a combination of trade agreements and domestic reforms in order to be effectively addressed.

The first step in estimating the economic costs and benefits of these measures was to estimate the trade costs of NTMs. There are two approaches to doing this: the quantitative approach using gravity models, or the price approach where direct questions are asked about the costs of the measures. The price approach was applied to the Japan study. The questionnaire sent to EU companies in Japan directly asked what the most important NTMs were and how much they cost in terms of additional costs in order to put products on the Japanese market. The questionnaires were sent to firms in the pharmaceutical, medical devices, processed foods, motor vehicles, transport equipment, and financial and telecommunications services sectors. These sectors were selected because these sectors registered a high number of NTMs and these sectors were important to EU-Japan trade.

The survey targeted 250 members firms of the European Business Council in Japan and with more than 100 firms responding we had a very good response rate. Most sectors sent back good replies, with the exception of financial and telecommunications services sector firms, as it was very difficult for those companies to give exact answers to those questions. In the end, the method of the survey had to be changed for these firms.

The pharmaceutical sector answered that the total cost of all NTMs amounts to 22% of additional costs in order for a European pharmaceutical manufacturer to put its product on the market in Japan. The causes and factors that create this cumulative 22% include the absence of a common list of additives, the introduction of new standards, questions related to the pricing and reimbursement system for medicine, and mutual recognition of conformity assessment. As for the additional costs, 20% out of the total 22% stem from measures that could possibly be reduced, though this may be an overly optimistic assumption. For medical devices, total additional costs were 30% and only 12% can be reduced. For processed foods, total additional costs were 25% and only 9% can be reduced. For motor vehicles, total additional costs were 10% and 9% could be eliminated. As for the service sector, because the initial survey was unsuccessful overall, a separate study was used where barriers were estimated on the basis of gravity models or quantity estimates.

After the trade costs were attained, they were plugged into general equilibrium models and the part of the trade costs that could be removed were taken out of the equation. Regarding the impact of removing NTMs on exports from the EU to Japan, if Japan removed all tariffs on processed foods, for example, the model estimates that this would increase EU exports by 4.8 billion euros, thus doubling EU food exports to Japan. If all the costs of NTMs could be removed through negotiations, another 1 billion euros could be added to this total. This is the most optimistic, and perhaps unrealistic, scenario. A minimalist reduction of NTMs shows that 0.3 billion euros would be added to exports from the EU. Processed foods and agriculture are the only sectors where there are significant remaining tariffs in Japan, which is why the tariff effect of removing those NTMs is very strong. The tariff effect is very small in the other sectors, and the most important reduction in other sectors comes from reducing NTMs. The increase that is expected from removing all tariffs in an FTA would be about 14 billion euros of increased EU exports to Japan. The impact that can be expected from removing all NTMs, however, is 29 billion euros, thus explaining why the EU has been insistent on addressing NTMs rather than tariffs in trade agreements.

Regarding Japanese exports to the EU, the impact of removing tariffs is somewhat stronger in most sectors, but is clearly the strongest in motor vehicles, where tariffs range from 11-14%. Removing those EU tariffs would increase Japanese motor vehicle exports to the EU by about 16 billion euros, thus almost doubling Japanese car exports to the EU. On top of that, removing NTMs on the EU side would add another 11 billion euros to exports.

It is important to note that in this study, NTMs were not only assessed on the Japanese side, but on the EU side. The level and the cost of NTMs in the EU were taken from another study conducted on EU-U.S. trade relations.

There has been much speculation as to the effect this would have on industry. The impact on output in the EU is negative in the car sector at -3%, but the output effect is very small in other sectors. In total, EU output would increase by 1.4%. For Japan, total production would increase by 0.03%, but there are some stronger effects in a variety of sectors, including the motor vehicles sector, which shows a strong positive effect, and also electrical machinery. Negative effects would be seen in sectors such as other machinery and agriculture.

Trade volume is something that companies are very interested in, but individuals are more interested in actual income increases. The model used also offers a simulation on expected impact on national income. For both the EU and Japan, and as a result of both the reduction of tariffs and NTMs, the results would be positive. The strong effect on income in the EU is related to the welfare effect. If tariffs on car imports in the EU are removed, for example, cars will be cheaper and the consumer can buy more cars for the same income, thus causing a real income effect.

It is important to note that these economic simulations were done under the assumption that there is no Doha Agreement. Factoring in a Doha Agreement, which itself lowers tariffs, would decrease the benefits from tariff reductions. Also of note is that the economic simulations were done without factoring in the EU-ROK agreement. This would have implications for the motor vehicles sector, since Korean car manufacturers would gain an advantage in access to the EU market, and a subsequent trade agreement with Japan would cause a substitution effect between Korean car exports and Japanese car exports.

There are a number of things in the study that could not be included in the model, but are still important from the perspectives of trade and economics. One of the most important factors that could not be included was public procurement. The study found that the size of public procurement markets in the EU and Japan are about the same in terms of percent of GDP, around 19%. The estimated size of the public procurement market in Japan in 2007 was about 565 billion euros. However, there is a gap between the levels of openness in both markets. In Japan, less than 4% of the total public procurement market is open to EU tenders, whereas in the EU, 70% of the public procurement market is open to Japanese tenders. If the Japanese public procurement market were more open, there would be enormous export gains for the EU.

Two other factors that were not taken account of by the model are railway equipment and aircraft. The aircraft market in Japan is about 4.5 to 5 billion euros per year. However, the EU only has about 10% of that market and the U.S. has 85%. Worldwide, however, the split between the EU and the U.S. in this market is 50/50. The two countries that do not share this ratio are Israel and Japan. Moving to this ratio in Japan for the aircraft market would cause EU aircraft exports to increase by about 2 billion euros. Similar reasoning can be applied to the railway equipment market in Japan, as the size of the market in Japan is about 2 billion euros per year and the EU's share of that market is less than 3%. However, in the worldwide railway equipment market, the EU has a 60% share, and getting to this level in the Japanese market would create large-scale gains for the EU.

Finally, an important issue is competition policy. There are no quantitative estimates for competition policy issues in Japan, but it has been found in a number of studies that competition policies and rules exist, but enforcement is weak, and thus ranks rather low on the OECD competition indicators. This is especially true in the service sector, where openness to import and internal competition is rather low since access costs are very high. The study has not come up with recommendations for how this could be dealt with via a trade agreement.

Q: First, has the same type of survey been conducted with South Korea and Canada, and if so, what did the results look like compared to the study on Japan? Second, it seems that, regarding exports, having no tariffs favors the EU and having tariffs favors Japan, though the results seemed to be mixed. Is this understanding correct?

Bertin MARTENS
Studies on South Korea and Canada did not go into the same amount of detail as the study on EU-Japan trade relations. They were conducted a few years ago and the methodology was based on the quantity approach. Also, detailed surveys were not conducted because this type of survey work is rather expensive to carry out.

On South Korea, studies have restarted since the EU-South Korea trade agreement because many questions have been raised about the motor vehicles sector in that regard. A team of experts on the motor vehicles sector have been hired to look into NTMs in that sector, but it is not available yet.

Regarding the interpretation of export increases for Japan and the EU, I agree with the interpretation that the results can be interpreted either to favor Japan or the EU. There are, however, more important issues involved here. For one, tariffs are easy to remove. NTMs, on the other hand, cannot simply be abolished since regulatory measures have a purpose. The question is to what extent the costs of those regulations can be reduced, and to what extent the trade costs of those regulations can be reduced. The report is based on the assumption that although there are differences in regulations between the EU and Japan, this does not mean to say that one is better than the other. It was observed that because of the difference in regulation between Japan and the EU, there are costs involved in trade between the two. The question is not about eliminating NTMs, but how far the trade costs of NTMs can be eliminated. The purpose of the study was to attach values to these NTMs and there is margin for debate as to the preciseness and validity of those values. The interpretation of the different methods of tariff removal and NTM reduction depends also on the validity of the methods that are chosen.

Q: What was the methodology used to decide whether in each sector there is more or less potential to reduce NTMs than in another sector?

Bertin MARTENS
The methodology was that the question was posed directly to firms in the surveys. In this sense, this study was a new kind of exercise. In the past, the methods that were used to address these issues were more statistical, econometric methods where trade data went through statistical analysis and tried to derive from that a figure for trade costs. That approach has its merits, but if this study was to be used in trade negotiations, a more down-to-earth method was necessary, which is why an empirical approach was decided upon. So far, no other studies can be found that have used this methodology to study trade costs.

Many of the trade costs involved cannot actually be addressed simply by trade policy instruments. Pricing and reimbursement, for example, is clearly a domestic policy issue and it is difficult to deal with it in trade negotiations. For this reason, the EU insists on seeing credible signs that Japan is moving forward in domestic reforms in order to create an environment in which the EU can be confident that domestic reforms will move forward.

Q: What is the relationship between this study and the assessment that the European Commission has been asked to conduct when they start trade agreement negotiations? Also, what is the difference between the methods of those sustainability impact assessments and this study?

Bertin MARTENS
In the EU, many types of studies take place. This study was meant to try to apply an economic value to these NTMs to get a clear picture of what is most important regarding trade. Regarding sustainability impact assessments, they are part of the EC's standard procedure, but they are undertaken only once approval to launch trade negotiations from the EC Council of Ministers is attained. This is because the most important part of a sustainability impact assessment is consultation with civil society on the potential impacts of such negotiations, which runs in parallel with the negotiation process. The EU and Japan are not yet at this stage, so it would be too early to start a sustainability impact assessment. Also, sustainability impact assessments would go further than this study in that they also include environmental and social issues.

Q: Regarding the output impact on the EU, though this study is a purely economic study, in reality, in the motor vehicle sector, Japanese companies are now establishing production sites in the EU, regardless of tariff conditions. It is unclear whether this has actually induced greater exports from Japan to the EU. It has been assessed by European car manufacturers that luxury car imports from Japan may increase, but the main market of cars will continue to be produced in Europe. How are realities like this for local companies reflected in the study?

Bertin MARTENS
This refers to substitution of investment flows and trade flows. The study attempts to address this by looking into a number of existing studies, but in the end it really could not be done. Many in the past have tried to build models that reflect the substitution between trade and investment and one of the key questions is to what extent trade and investment are complementary or substitutes for each other. Different research finds different answers to this question, though the answer depends highly on the sector that is being studied. In the case of the motor vehicle sector, what was found in a review of literature was that there is no consensus on this point. It is also difficult to make estimates since the motor vehicle sector is seeing worldwide difficulties with the current economic climate.

Q: Regulatory dialogues are ongoing between Japan and the EU, and while they are useful, they are one-off meetings rather than negotiations. To put the maximum pressure on the regulators concerned, a negotiation framework is needed. While a negotiation framework for FTAs and EPAs is in place, that framework needs to be utilized. Related to this, though Japan has many problems in the service sector, Japan is open to discussions on this issue. In addition, Japanese businesses have shifted their views toward NTMs along with tariff issues.

Bertin MARTENS
The EU has also tried to assess its reform dialogue history with Japan and there has been much internal debate within the EU on what an appropriate format would be for these dialogues. Some believe the dialogue should move to an FTA setting. The study shows that many of the issues discussed in the regulatory reform dialogue were not purely domestic reform issues. Some of the issues, though not all, can be captured in an FTA. However, while FTAs might take a top-down approach, there is much in the way of bottom-up approaches in terms of regulation. The EU has been signing trade agreements mainly with developing countries, and South Korea and Canada are the first two OECD countries with whom the EU has engaged in trade negotiations. Trade negotiations on standards among the EU, Japan and the US would be a very different kind of negotiation.

Comment:

Standards issues are very technical in nature, but a negotiation framework would allow for the development of new ideas. If the EU and Japan could enter discussions on this point, they could begin building a future framework for discussing NTMs. Many possibilities are being discussed for setting up a framework for fixing international standards.

Comment:

The Japanese government is fully committed to regulatory reform and a minister promoting EPAs and FTAs asked whether Japan should push forward trilateral harmonization of standards and regulations. Some areas of harmonization should be promoted, but other areas are more focused on this in regards to the EU.

Regarding railways, there has been a request from European companies to sell railway equipment and systems in Japan. One European company did sell a system to Japan, but needed to rely on Japanese management to run the system. It is difficult for European companies to achieve the same success as the Shinkansen.

Q: The sample used in the survey might underestimate NTMs due to selection bias. Are the companies in the survey happy with reducing NTMs as they have already invested in overcoming NTMs in Japan?

Bertin MARTENS
European companies would be very interested in reducing the costs of trade through NTMs since they have been lobbying the EU strongly to push the Japanese authorities to come to an agreement on this. Some companies may be interested in market segmentation and may want to retain differences in regulation. However, no examples of companies that explicitly stated that they prefer greater market segmentation were recorded in the survey.

Regarding selection bias in the sample, there may be some, but the sample of companies cover the most important export sectors for the EU. All in all, the sample is reasonably good.