At this stage, the Commission has at its disposal sufficient evidence that imports of the product concerned from Turkey are being dumped. The complaint provided sufficient evidence of dumping based on a comparison of the normal value thus established with the export price (at ex-works level) of the product concerned when sold for export to the Union. As a whole, and given the extent of the alleged dumping margins ranging from 4 % to 8 %, this evidence provided sufficient support that the exporting producers practice dumping.

The complaint also provided sufficient evidence of alleged injury to the Union industry, including a negative development of key performance indicators of the Union industry.

That information was contained both in the non-confidential version of the complaint and in the Notice of Initiation published on 14 May 2020. By its publication in the Official Journal of the European Union, the Notice of Initiation is a public document accessible to all importers. Furthermore, as interested parties in the investigation, importers have access to the non-confidential version of the complaint and the non-confidential file. Therefore, the Commission considered that, on this basis, importers, were aware, or should have been aware, of the dumping, the extent thereof and the alleged injury3.

As stated in recital 5 several interested parties submitted comments to the request for registration. With regard to the first criterion, the Government of Turkey, Habas, CIB and the Colakoglu and Erdemir Groups highlighted that the mere initiation of an anti-dumping investigation does not automatically mean that dumping is actually taking place, as certain investigations (some concerning Turkey) are terminated without the imposition of duties. Rather it would be only a ‘one-sided allegation’. Thus, an importer cannot be aware of something (‘dumping practices’) that the investigation has not yet established. The Government of Turkey, Habas and CIB also contested certain additional evidence provided by Eurofer in the request, including the claim that the CEO of a Turkish exporting producer was aware of the upcoming anti-dumping investigation and the information that imports of the product concerned are subject to trade defence measures in third countries. Lastly, TCUD noted that some of the claims made by Eurofer regarding anti-dumping measures against Turkish imports of the product concerned were not truthfully described in the submission. In this respect, TCUD argued that one company was exempted from the US measures following a challenge before the relevant courts and that another third country (Morocco) removed the anti-dumping duties against Turkey following the ruling of a WTO panel.

Article 14(5) of the basic Regulation stipulates that, as of the initiation of the investigation and having informed the Member States in due time, the Commission may direct the customs authorities to take the appropriate steps to register imports, so that measures may subsequently be applied against those imports from the date of such registration. The main objective of registration in this case is to allow the possibility to retroactively impose provisional measures up to 90 days preceding their application in accordance with Article 10(4) of the basic Regulation. In light of this provision, the Commission cannot register imports after a provisional duty has already been imposed.

If, as suggested by some parties, registration is only possible after a finding of dumping has been made by an investigation, then by definition it could never take place prior to the imposition of provisional duties. Such approach would completely deprive the registration tool of its effet utile4. Therefore, the Commission needs to take a decision on whether to register imports or not before any provisional duty is in place. In this respect the Commission noted that at the stage when the complainant filed the request for registration the best information available was that of the complaint, on the basis of which, as provided in the Notice of Initiation, the Commission initiated the anti-dumping investigation, and which indicated the existence of evidence tending to show injurious dumping. Therefore, the Commission rejected these claims as unfounded. Moreover, in view of this finding, the Commission did not need to assess the additional evidence referred to above regarding imports of the Turkish exporting producers to third countries and the related arguments.

The Commission thus concluded that the first criterion for registration was met.

The Commission analysed this criterion based on the statistical data for the product concerned available in the Surveillance 2 database. For assessing whether a further substantial increase had taken place since the initiation of the investigation, the Commission first defined the periods of time to be compared. On the one hand, it assessed the import data from Turkey following the initiation of the anti-dumping investigation (i.e. the point in time as of when importers were aware, or should have been aware, of dumping practices) until the most recent period, i.e. the period June-mid-October 2020. On the other hand, the Commission calculated Turkish imports for the same period during the investigation period5 (i.e. June-mid-October 2019) and the monthly average import volumes in the full investigation period.

The Commission considered that, in this case, a comparison between the post-initiation average monthly import volumes with the average monthly Turkish import volumes in the entire investigation period would not be sufficient to assess whether there was a substantial increase in imports in view of the market developments and stockpiling effect derived from the ongoing safeguard measures affecting the product concerned. Thus, the Commission considered it appropriate to base its determination on the comparison between the post-initiation average monthly import volumes with the same period in the investigation period in order to better reflect the potential influence of the ongoing safeguard measures on the behaviour of import flows from Turkey in the period assessed6. Comparing the post-initiation import volumes with import volumes in the same period of the previous year would, in addition, take due account of possible seasonality effects, if any7.

Regarding the influence of the safeguard measures on the Turkish import flows, the Commission noted that throughout the whole year 2019 (the investigation period), Turkey’s imports of the product concerned were affected by numerous adjustments to the functioning of the safeguard measures that impacted the export behaviour from Turkey.

First, January 2019, by far the best-performing month of Turkish imports of the product concerned, coincided with the last month of provisional safeguard measures. In this respect, there was uncertainty in the market as to the shape definitive measures would take. The Commission noted that any volume of free-of-duty Tariff-Rate Quota (‘TRQ’) not sold by the end of that month would not have been carried over to the next period, i.e. volumes of free-of-duty TRQ would have been ‘lost’. A similar situation, i.e. last month(s) prior to a new period of the measures, occurred in the quarter February-March 2019. Second, in February, April, July and October of the year 2019 there were openings of new free-of-duty TRQ batches. Under the safeguard measures on certain steel products, every opening of TRQ has generally lead to large volumes being imported in the early stages of each of the quarters. Last, the Commission announced in August 2019, in the framework of the first review investigation of the steel safeguard measures, that it would make Turkish imports of the product concerned subject to a 30 % cap over the TRQ available under the relevant product category, thus also leading to an increase in imports ahead of such change in the TRQ administration.

The Commission further highlighted that the above events impacted the flow of import volumes as shown in the statistics. Notably, the months of January, March and September rank first, second and fourth in terms of volumes exported monthly by Turkey in 2019 (in some cases being substantially higher than any other month in that year)8. Hence the Commission concluded that all these regulatory actions in the framework of the steel safeguard measures influenced significantly the behaviour of market operators and the trend of import flows across the year.

The situation in a substantially shorter period (between June and mid-October) was less affected by regulatory changes in the management of the TRQs. Thus, the Commission compared the monthly average Turkish import volume in the period June-mid-October 2020 with the monthly average volume of imports in the same period of the preceding year, 2019.

Source: Surveillance 2 database.

The Commission further examined the import trends in the selected period and the respective period within the investigation period in order to determine whether that 6 % increase constituted a substantial increase in imports in this case. In particular, the Commission observed that negligible import volumes from Turkey took place in the month of September 2020. Such exceptionally low volume of imports was largely due to the fact that under the steel safeguard measures, Turkey had virtually exhausted its country-specific TRQ in that month in product category 1 (where the product concerned belongs) and therefore it would have been only able to export any meaningful amounts of the product concerned subject to a 25 % duty10. This exceptionally low import volume in September 2020 reduced to a large extent the monthly average import volumes in the relevant period chosen by the Commission to carry out the comparison (June-mid-October 2020). At the same time, the Commission noted that precisely in the same month of the preceding year, in September 2019, there was a very high volume of imports due to the fact that in October 2019 the adjustments following the first review investigation of the safeguard measures would take effect. Therefore, in anticipation of this event, Turkish producers exported very large amounts during the month of September 2019.

In light of the considerations in the previous recital, the Commission considered that the imports of the months of September of, respectively 2019 and 2020 were not representative and assessed also the import levels without including that month in both years. Such comparison confirmed that the increases of import volumes in comparison with the same months during the investigation period had been very significant. For example, when comparing the period June-August in 2020 and the same period in 2019 there was an increase by 44 %. A significant increase by 49 % would also take place when comparing the period June-mid-October 2020 with the same period in 2019, in the absence of the month of September. In light of these considerations, the Commission concluded that the increase of imports in this case is substantial.

Regarding this criterion, the Government of Turkey, Habas, CIB, the Colakoglu and Erdemir Groups and the Consortium of HRFS users contested the period proposed by the complainant to assess whether a substantial rise in imports took place. Instead, these parties proposed different periods to carry out such comparison that in their view would show that this criterion was not met. Habas and the Consortium of HRFS users suggested that comparing on a quarterly basis the same periods of 2019 and the first half of 2020, i.e. the first two quarters of each year, would show a decrease in imports. CIB and the Colakoglu and Erdemir Groups argued that the period chosen by Eurofer, namely June-July 2020, was too short, and thus not representative. Moreover, CIB and the Consortium of HRFS users alleged that the reasons for any increase in imports would have nevertheless been the consequence of the relaxation of the lockdown measures in the Union related to the COVID-19 pandemic, which coincided with the early stages of the post-initiation period. CIB argued that the Commission must compare the volume of imports that occurred after initiation of the investigation with the volume of imports that occurred during the investigation period. This party further argued that Turkey’s imports would have shown a steady decline in January-May 2020 and that in the months of June and July 2020, they would have just increased to return to ‘their normal level’. Moreover, CIB concluded that the Commission should disregard the post-initiation data of June 2020 because of the impact of the alleged particular ‘post-quarantine’ situation. In such case the comparison with the monthly average in the investigation period would show an increase of 12 %, which could not qualify as ‘further substantial increase’, but rather as a ‘very modest’ increase. Other parties proposed to assess the import trends in the first seven months (January-July) of the periods 2017-2020, the import trends throughout the years 2017 to 2020 (annualised)11, the monthly average in the years 2018 and 2019 with the monthly average in the period January-July 2020, the monthly average of the period January-July 2018 and 2019 with the same period in the year 2020, and the monthly average in the investigation period with the import volume in the month of July 202012, to show that there would have been a decline in imports. The Colakoglu and Erdemir Groups also took issue with the fact that the request did not include any import data between the end of the investigation period and the initiation of the investigation, despite the time lapsed between the two.

Furthermore, the Consortium of HRFS users additionally claimed that registration would reduce the level of imports and cause a shortage of supply which would be threatening the operations of certain users’ production facilities established in the Union. This would allegedly disrupt the market and cause irreparable damage to independent users of the product concerned.

All interested parties that reacted to the registration request from Eurofer concurred in arguing that because of the impact of the steel safeguard measures on Turkey, it would not be able to rise further its exports in the coming months (i.e. those prior to the imposition of provisional anti-dumping measures), and that therefore the Commission should disregard the request for registration.

With regard to the different periods proposed by the parties, the Commission refers to the reasoning and explanations it laid out in section 3.2 above. Nevertheless, there are a number of claims in this respect that warrant a specific rebuttal. In the first place, the Commission notes that while it usually also compares the monthly average imports of a post-initiation period with the average import levels in the investigation period, such comparison may not always be appropriate. In this case, the Commission has found and explained in recitals 16 to 20 above that, in view of the particular market circumstances with regard to the imports under investigation, using the average imports of the year 2019 is not the most appropriate basis for comparison.

Second, the Commission notes that one of the proposals referred to trends from a period prior to the initiation of the anti-dumping investigation. This would be inconsistent with the finding that the awareness of dumping practices from Turkey stems from the publication of the Notice of Initiation. Thus, the trend in imports from a period prior to the point where an importer could have been aware of the existence of dumping, is not to be taken into consideration for assessing the merits of a registration request.

Third, the Commission further notes that, while Eurofer’s request constitutes the starting point for the analysis, the Commission nevertheless also has to carry out a comprehensive analysis of its own. In this regard, the Commission has at its disposal a set of data for a more recent period than that provided by Eurofer in the request (namely, until mid-October 2020)13. Moreover, the Commission also deemed it more appropriate to compare this set of data with the same period in the investigation period, for the reasons developed above.

Thus, the Commission does not consider that the different proposals have shown any compelling reasons that would render them more apt to assess the import trends than that developed in section 3.2. Therefore, the Commission rejected these claims.

Regarding the claims that because of the safeguard measures in place, there would in any event be no risk of a further substantial rise in imports, the Commission disagreed for the following reasons.

First, the registration does not prejudge a decision on retroactive collection of the provisional anti-dumping duties. Whether safeguards measures will prevent an increase of imports until the imposition of provisional measures should be assessed in the context of a possible retroactive collection of duties14.

In any event, the Commission acknowledges that, as a result of the changes introduced by the safeguard measures as of 1 July 2020, Turkey has become subject to a country-specific TRQ whose quarterly volumes are set out in Annex II of Commission Implementing Regulation (EU) 2020/89415. If these volumes are exhausted, Turkey is nevertheless still allowed to continue exporting into the Union16.

However, the Commission does not agree that this fact alone would automatically prevent Turkish imports from rising substantially before the imposition of provisional anti-dumping duties, if any. This is so because by the time provisional duties may be imposed at the latest, this is by mid-January 2021, two batches of free-of-duty country-specific TRQ would have been released for Turkey17. Moreover, any eventual comparison between import volumes to assess whether retroactive collection of duties is warranted could also cover a period prior to that when Turkey became subject to a country-specific TRQ as explained in recitals 18 and 19 above. Therefore, at this stage, there is no evidence that the overall volumes that Turkey could export free-of-duty (and under the 25 % duty) until an eventual imposition of provisional anti-dumping measures could not qualify as a further substantial rise. The Commission will any event only be able to conclude on this matter in case definitive anti-dumping duties were imposed at the end of the ongoing investigation.

Therefore the Commission rejects the claims that because of the safeguard measures in place a further substantial rise in imports should be ruled out.

Lastly, the Commission notes that the comment made by the Consortium of users regarding the impact of registration on the supply of the product concerned and on independent users does not fall under the legal requirements that are to be assessed when deciding on the registration of imports. Therefore the Commission does not address it at this stage of the proceeding.

In view of the above considerations, the Commission found that the second criterion for registration was also met.

The Commission has at its disposal sufficient evidence that additional injury would be caused by a continued rise in imports from Turkey at further decreasing prices.

As established in section 3.2 there is sufficient evidence of a substantial rise in imports of the product concerned.

Source: Surveillance 2 database.

The further rise in imports following the initiation of the case is thus likely, in light of its timing, volume and other circumstances, such as the pricing behaviour of exporting producers, to seriously undermine the remedial effect of any definitive duty, unless such duty would be applied retroactively.

The Commission therefore concluded that the third criterion for registration was also met.