EFFECTS OF NEW US AUTO TARIFFS ON GERMAN EXPORTS, AND ON INDUSTRY VALUE ADDED AROUND THE WORLD

Transcription

1 1 ifo Institute ifo Center for International Economics Gabriel Felbermayr & Marina Steininger Feb 15, 2019 EFFECTS OF NEW US AUTO TARIFFS ON GERMAN EXPORTS, AND ON INDUSTRY VALUE ADDED AROUND THE WORLD The US are pondering additional tariffs on car imports due to concerns over national security. In this analysis, we study the effects of a permanent additional import tariff of 25% on car imports from all countries. Amongst the EU countries, y is by far most strongly affected by potential new US tariffs on car imports. If the US imposes permanent tariffs of 25%, in the long-run, car exports to the US could fall by almost 50%, i.e., by 17.0 bn. Euro. Total car exports would fall by 18.4 bn. Euro or 7.7%. Exports from other sectors and to other countries would compensate these effects slightly, so that the total loss of exports would amount to about 11.6 bn. Euro. Value added in the car industry goes down by about 7 bn. Euro or almost 5% of the initial level. Value added in the US car industry, by contrast, would go up by about 25 bn. Euro. The EU can devise a clever retaliation strategy that would approximately nullify the negative effects of US car tariff for both parties, but harm third countries. Auto tariffs ahead No later than Sunday February 17, the US Department of Commerce is expected to present the results of an investigation under Section 232 under the Trade Expansion Act of 1962 into the security threats of car imports into the United States of America. It is widely expected that the report will suggest that imports of cars, car parts and possible other motor vehicles poses a threat to American security interests. The US president will then have time until May 18, 2019 to make a decision whether or not to follow a recommendation to impose new tariffs on car imports. At this point, it is not clear how high these tariffs would be; a plausible tariff to assume is one of 25%. It is

2 2 also unclear, whether such a tariff would apply to all trade partners or only to a selection. The revised trade agreement between the US, Canada and Mexico comes with a side letter that shields the partners from possible Section 232 tariffs. However, that agreement is not ratified yet, and it is unclear whether this can happen until May 18. In this note, we ask: What would be the effects of erga omnes auto tariffs on the US economy and on trade partners? To provide answers, we use a relatively standard general equilibrium trade model, calibrated to data from 2014, and simulate the effects of the introduction of an additional import tariff on cars of 25%. As a consequence, cars would be subject to a total tariff of 27.5% in the US, as the initial tariff is 2,5%. The model, described in detail in various publications, the latest being Felbermayr et al. (2019) 1, covers 44 countries and 50 goods and services sectors. The Ricardian trade model features tariff and non-tariff barriers to international trade as well as a careful representation of the international and intranational input-output relationships. 2 This is of enormous importance for the case of car tariffs, because the automotive industry uses a lot of inputs supplied by other domestic or foreign sectors. All results reported in this note refer to the long-run, i.e., after all adjustment to the new tariff situation has taken place. For example, new US tariffs imply that car manufacturers will increase production in the US at the expense of domestic production; but this will entail increased exports of auto-related services or other inputs that are not affected by tariffs. Moreover, car manufacturers will try to find new markets; they will also have to adjust their pricing strategies. All these adjustments take some time. Typically, adjustment will be completed by 90% in five years and by 99% in ten years. Moreover, new tariffs are assumed permanent in the model. This is an important assumption because expectations about the durability of the new tariff regime strongly affect firms sourcing and pricing decisions. Finally, it is important to note that all results pertain to national economies; if car firms generate value added in the US, domestic income is affected only to the extent that foreign profits are repatriated. 1 Felbermayr, G., F. Kimura, T. Okubo, M. Steininger (2019), Quantifying the EU-Japan Economic Partnership Agreement, Journal of the Japanese and International Economies, forthcoming. 2 The data covers more than 90% of the worldwide input-output linkages.

3 3 Initial situation Table 1 reports the baseline situation for our study. The four countries with the largest direct car exports to the US are Mexico, Japan, Canada, and y. For y, exports worth about 34 bn. Euro are affected; this is 30% of total exports to the US. The second most affected car exporter in the EU is UK with almost 5 bn. Euro. Relative to total exports, Hungary and Slovakia are strongly affected. These countries host Audi factories, which export to the US. However, most EU member states are not strongly affected by new tariffs on autos. Note that the numbers in Table 1 do not include indirect exports: for example, Austria may export car parts to y, where they are assembled to full cars and exported to the US. The model takes those indirect exports into account. Trade effects for y Tables 2 and 3 show the effects of additional US import tariffs of 25% on auto trade (columns (4) and (5)). The simulations suggest that car exports to the US could go down by as much as 17 bn. in the long run; this is a drop of about 50%. Overall car exports would go down by even more, 18 bn. Euro. Disappointingly, the direct trade destruction effects are not mitigated by indirect trade diversion effects. There are two reasons for this: (i) The new car tariffs force the car industry to lower prices and this negative terms-of-trade effect spills over to other markets; so car manufacturers margins fall in all major markets. (ii) y is the center of the European car production network. It sells car parts to countries such as Spain, Hungary or Slovakia, where cars are assembled. imports of cars fall, too, mostly because demand for car parts in y goes down. This is also true for domestic sales. In the baseline, y reports car sales of 379 bn. Euro; 63% of which (239 bn. Euro) are exported. US tariff reduce car sales by 4.9% and car exports by 7.7%. Total exports to the US go down by about 13.5 bn. Euro. This is less than the effect on car exports alone, because the relative shrinkage of the car industry sets free resources, which will be employed in other sectors, and these increase their exports to the US. Global exports (i.e., to all countries in the world) fall by 11.6 bn. Euro, which is still less, as y s exports to other countries (in non-auto sectors) increase.

4 4 Value added in the automotive industry According to the simulations, the new US tariffs would depress value added in the car industry by about 7 bn. Euro; this is about 4.8% of the initial baseline value. Japan and Mexico, which have substantially higher car exports to the US, lose about 8 bn. Euro each, 9% and 21% of initial levels, respectively. Canada would suffer a loss of about 5 bn. Euro, which would, however, amount to 31% of the initial level. However, as noted above, both Canada and Mexico could be exempted from the tariffs (but would have to accept quantitative restrictions). In this case, car manufacturers would suffer even more as they might lose additional market share in the US to Mexico and Canada. In Europe, the British car industry would suffer a loss of about 0.8 bn. Euro, followed by Italy with a loss of 0.6 bn. Euro. In contrast, value added of the US car industry would go up by about 25 bn. Euro or 18% in the long run. This comes at a cost for the entire world: the global automotive industry would lose about 6 bn. Euro. The rest of the world loses more than the US win. Effects on real income Finally, Tables 2 and 3 show the economy wide effects on GDP, taking into account tariff income. The effect of tariffs is very important: in the initial situation, the US has car imports worth about 216 bn. Euro; after the tariffs, imports still amount to 108 bn. Euro. Thus, the US collect duty revenue worth 27 bn. Euro on this sum. In y, the new US tariffs would shrink GDP (real, i.e., price-adjusted income) by about 5 bn. Euro, as ifo has reported last year. 3 Thus, it would bear 57% of the EUwide damage. This number includes demand linkage effects from car parts suppliers. Mexico, Canada, or Japan would lose slightly less than y. The US, in contrast would add about 5 bn. Euro to its price adjusted GDP. Should the EU retaliate? The EU has announced that it would react with balancing tariffs should the US impose car tariffs on EU countries. Figure 1 shows the effects on real income when the US imposes auto tariffs and tariffs on steel and aluminum (already in place). As seen above, this generates a loss of about 5 bn. Euro for y (the effects of steel and 33 See the press statement from May 24, 2018 at

5 5 aluminum tariffs are very small). If the EU retaliates by choosing those imports from the US where tariff incidence is most favorable (i.e., where US producers are most likely to lower their export prices), it can compensate the GDP losses. The global economy, however, would suffer from such a tariff war. For the US, the initial advantage of the tariffs falls to a negligible 1.6 bn. Euro. This exercise suggests that in case the US attacks Europe with car tariffs, it is sensible for Europe to retaliate, but the global economy would suffer damage. Table 1 Automotive and overall exports to the US in the baseline (2014), bn. Euro. Automotive Overall Automotive Overall MEX POL JPN AUS CAN ROU DEU NLD KOR PRT CHN CHE ROW FIN GBR RUS ITA SVN TWN DNK AUT NOR ESP IRL HUN BGR IND EST SWE HRV FRA LUX SVK GRC BEL MLT BRA LVA TUR LTU CZE CYP IDN Note: ROW denotes rest of the world. Data come from the WIOD project whose sector definitions follow standards in Input-Output tables so that numbers are not identical to those from national or other sources.

6 6 Table 2 US car tariffs: Effects in EU countries and on trade flows with EU countries Economy-wide Changes Exports Imports Changes in Automotive Sector Exports Imports Real Income Value Added (1) (2) (3) (4) (5) (6) DEU AUT BEL BGR CYP CZE DNK ESP EST FIN FRA GBR GRC HRV HUN IRL ITA LTU LUX LVA MLT NLD POL PRT ROU SVK SVN SWE EU Source: Own calculations based on the ifo trade model. EU28 Average for Real Income change, sectoral value added includes y, while EU28 average for the trade changes excludes y. Long-run effects.

7 7 Table 3 US car tariffs: Effects in non-eu countries and on trade flows with non-eu countries, mn. Euro Economy-wide Changes Exports Imports Changes in Automotive Sector Exports Imports Real Income Value Added (1) (2) (3) (4) (5) (6) AUS BRA CAN CHE CHN IDN IND JPN KOR MEX NOR RUS TUR USA Source: Own calculations based on the ifo trade model. EU28 Average for Real Income change, sectoral value added includes y, while EU28 average for the trade changes excludes y. Long-run effects.

8 8 Figure 1 Real income effects of US Section 232 tariffs with and without retaliation, bn. Euro. USA Austria China Global USA unilateral Retaliation EU y EU Source: Own calculations based on the ifo trade model. EU28 Average for Real Income change, sectoral value added includes y, while EU28 average for the trade changes excludes y. These calculations include all Section 232 tariffs (car tariffs plus steel & aluminum tariffs). Long-run effects.