Analytical Note

China Economic Insights

Macroeconomic Impacts of Trump 2.0 Tariffs on China: A Model-

Based Assessment1

April 19, 2025

____________

I. Introduction

1. The trade war initiated by the Trump administration in 2018 marked a turning
point in U.S.-China economic relations, underscoring broader global trade conflicts.
With longstanding allegations against China of unfair trading practices, intellectual property
theft, and forced technology transfers, the first Trump administration initially imposed tariffs on
a wide range of Chinese goods, starting at US$50 billion and eventually encompassing a total
of US$360 billion. China retaliated with its own tariffs on U.S. imports, escalating the conflict
into a full-blown trade war. This led to disruptions in global supply chains, diminished bilateral
trade volumes, and heightened businesses uncertainty. Consumers in both the U.S. and China
faced increased prices due to the tariffs. The phase-one trade deal which was signed in early
2020 halted further tariff hikes, but the existing ones were left unchanged even under the
Biden Administration, and exacerbated with new tariffs and other protectionist measures.

2. With Trump’s second term, the world faced a renewed wave of protectionism
and anti-globalization backlash. President Trump has vowed to impose higher tariffs and
stricter import restrictions to reduce reliance on China and to protect and develop industries
in the US. Upon assuming office in January 2025, he announced tariffs of 20 percent on China,
25 percent on Canada and Mexico, and 25 percent on steel, aluminum, and automobile
imports. On April 2, 2025, the Trump administration officially unveiled a reciprocal tariff policy,
signaling a shift toward broader and more aggressive protectionism. In response, China has
intensified its retaliatory stance, announcing a 34 precent tariff on selected U.S. imports—a
sharp escalation from its earlier, more restrained reactions. 2 The U.S. responded on April 8th
by imposing an extra 50 percent tariff on all imports from China, bringing the total cumulative
tariff increase on Chinese goods to 104 percentage points. On April 9, China countered by
increasing its tariff on U.S. goods to 84 percent. President Trump responded by announcing

1 Prepared by Fan Zhai (zhai.fan@amro-asia.org) (Country Surveillance); reviewed by Jae Young Lee (Group
Head); and authorized by Hoe Ee Khor (Chief Economist). The views expressed in this note are the author’s
and do not necessarily represent those of the AMRO or AMRO management. The author thanks Xianguo Huang
for helpful suggestions and comments.
2 In response to the two rounds of 10 percent U.S. tariff hikes in February and March 2025, China imposed tariffs

of 10–15 percent on U.S. coal, LNG, crude oil, agricultural machinery, selected agricultural and food products,
and certain automobiles. These measures raised China’s average tariff rate on U.S. imports by just 2.5
percentage points in total.
a further 41 percentage point increase in tariff rate on China to 145 precent. 3 On April 11,
China increased its retaliatory tariffs on U.S. imports to 125 percent. This sequence of tit-for-
tat measures reflects rapidly rising tensions and underscores the growing risk of a full-scale
U.S.-China trade war.

3. This note presents a model-based analysis of the potential impact of tariff

policies under Trump’s second administration. It employs a multiregional, intertemporal
dynamic general equilibrium model of the world economy to simulate different scenarios for
the Trump 2.0 trade war. The modeling exercise attempts to answer the following questions:
What are the macroeconomic impacts of the Trump 2.0 tariffs on the Chinese economy? What
are the effects of China’s retaliatory tariffs against the U.S.? How would fiscal stimulus by the
Chinese authorities help mitigate the impacts of trade war? The simulation results suggest
that Trump’s protectionist trade policies would impose substantial economic costs on the U.S.,
China, and the global economy. They also highlight the importance of exchange rate flexibility
as a shock absorber and the role of expansionary fiscal policies as a stabilization tool for China
to mitigate the tariff headwinds.

II. Model

4. The model employed in this study is a version of a multi-country dynamic

general equilibrium model for the world economy, inspired by the new open economy
macroeconomics literature. It combines the long-run properties of neoclassical models with
short-run dynamics arising from nominal rigidities à la new Keynesian macroeconomics. The
structure of the model closely follows the global integrated monetary and fiscal model (GIMF)
developed by the International Monetary Fund (IMF) (Kumhof et al. (2010); Anderson et al.
(2013)). Agents in the model are forward-looking, endowed with perfect foresight and subject
to dynamic budget constraints. The model features overlapping generations with agents
having finite economic lifetime. This leads to the non-Ricardian feature of the model and
makes it suitable for fiscal policy analysis. Countries and regions in the model are linked
through trade and financial markets. Nominal price and wage stickiness, as well as real
frictions in investment, are incorporated to generate more realistic adjustment dynamics. The
presence of nominal price and wage rigidity allows monetary policy to play a key role. Unlike
the GIMF, our model is deterministic, excluding stochastic shocks or other uncertainties. The
model is set at an annual frequency and calibrated to the world economy of 2023 using macro,
trade and production data. 4

5. The world economy in the model is segmented into three economic blocs: the
US, China, and the rest of the world (ROW). There are four agents in each region, namely,
households, unions, firms, and government. Households have finite lives, facing a constant
probability of survival, as in the perpetual youth model consistent with Blanchard (1985) and
Yaari (1965). Households consume a basket of goods and services and exhibit habit
persistence in their consumption. The model distinguishes two types of households: forward
looking ones and liquidity-constrained ones. The former own a portfolio of domestic firms.
They also hold two types of nominal bonds: domestic bonds issued by the domestic
government denominated in domestic currency, and international bonds issued by the U.S.
and denominated in U.S. dollars. International bonds are traded only bilaterally with the U.S.

3 At the same time, Trump announced an immediate 90-day pause, during which all countries—except China—will
have their reciprocal tariff rates temporarily set to the minimum 10 percent.
4 Detailed specifications of the model are described in Appendix A and key model parameters are reported in

Appendix B.

2
and issued in zero net supply worldwide. The liquidity-constrained households do not have
access to domestic or international capital markets. They finance their consumption
exclusively with current disposable labor and transfer incomes.

6. The production sector features a continuum of firms that produce differentiated

varieties of products. Similarly, each economic bloc features a continuum of unions that
purchase labor services from households and supply labor to firms. 5 These unions act as
monopolistic suppliers of differentiated labor inputs to domestic firms and face nominal
rigidities in wage-setting. They set nominal wages according to constant elasticity downward-
sloping demand schedules and quadratic costs of wage adjustment as in Rotemberg (1982).
Production activity is characterized by monopolistic competition. Firms set the nominal prices
of their products in domestic and export markets to maximize the present discounted value of
profits. Similar to wage setting, price changes are subject to adjustment costs, which give rise
to nominal price rigidities. Firms’ investment is subject to adjustment costs, which allow for the
variation in Tobin’s q and generate plausible investment dynamics.

7. Firms are assumed to follow dominant currency pricing (DCP) when setting
prices in the export market. International trade is modeled using a nested Armington
structure. Domestic absorption is divided between domestic goods and aggregate imports,
with aggregate imports further allocated across sourcing countries, determining bilateral trade
flows. Demand for domestic and imported goods is expressed as a composite good defined
by the Dixit-Stiglitz aggregator over domestic and imported varieties. Firms set export prices
in terms of the U.S. dollar, i.e., traded goods are invoiced in a dominant currency (Gopinath
and Itskhoki, 2022). This specification leads to an asymmetric exchange rate pass-through to
the import and export prices of non-U.S. regions. Drawing on empirical evidence from the
2018-19 U.S.-China trade war (Cavallo et al. (2021); Fajgelbaum and Khandelwal (2022)), 6
we assume that import tariffs are excluded in the price-setting equation of exporters, i.e., any
changes in tariffs are fully passed through to consumers.

8. Governments in the model are characterized by their fiscal and monetary

policies. Governments collect taxes and issue debt to finance their budget deficits. There are
five types of tax in the model: labor income tax, capital income tax, sales tax, import tariff, and
lump-sum tax on households. Government consumption and investment are exogenous and
the lump-sum tax on households is endogenously adjusted to achieve a target path for the
desired government debt-to-GDP ratio. The monetary policy rule in the model follows a
Henderson-McKibbin-Taylor rule in which the nominal interest rate depends on the lagged
nominal interest rate, the inflation gap and the output growth gap.

III. Impacts of Trade War

9. We model the implications of the Trump 2.0 tariffs and China’s retaliatory
measures through three scenarios. The first scenario—Trump tariffs scenario—simulates
the effects of all tariff measures implemented by President Trump in his second term up to the
April 2 reciprocal tariff announcement. Goldman Sachs (2025) estimates that the weighted

5 The introduction of unions is for model simplification, as aggregation across generations would be difficult if
nominal rigidities were faced by households rather than unions.
6 Based on U.S.-China trade and price data from 2018 to 2020, Cavallo et al (2021) found that a 10 percent tariff

was associated with a 0.6 percent decrease in ex-tariff price and a 9.4 percent increase in the overall price paid
by the importer in the first year in the U.S. market. In contrast, they found a much lower exchange rate pass-
through, with a 10 percent depreciation of the U.S. dollar leading to an approximately 2.18 percent increase in
import prices.

3
average tariff rate under the U.S. reciprocal tariffs is 18.3 percent. After accounting for product
exclusions—such as goods already subject to, or expected to be subject to, sectoral tariffs—
the effective rate of reciprocal tariffs is reduced to 12.6 percent, comprising 26.3 percent for
China and 10.5 percent for ROW. Incorporating earlier tariffs imposed on China, Canada, and
Mexico, as well as the tariffs already applied to steel, aluminum, and autos, this scenario
implies tariff increases of 47.5 and 14.3 percentage points on U.S. merchandise imports from
China and the ROW, respectively. The second scenario, labeled “China retaliation,” builds on
the first by incorporating China’s countermeasures. In this scenario, China imposes a
retaliatory tariff of 36.5 percent on its merchandise imports from the U.S., reflecting the
cumulative retaliatory tariffs enacted by China prior to April 5. The third scenario—escalated
trade war scenario—extends the second scenario by including additional rounds of retaliatory
tariffs exchanged between the U.S. and China between April 8 and 11. This includes a 91
percent increase in tariffs on both sides, resulting a total tariff increase of 145 percent by the
U.S. on Chinese goods and 125 percent increase by China on U.S. goods. In all scenarios,
tariff increases are assumed to be permanent.

10. The Trump tariffs lead to output losses and temporary inflationary forces in the
U.S. Compared to the baseline, U.S. real GDP declines by 0.15 to 1.2 percent in the first three
years following the imposition of higher tariffs in the Trump tariffs scenario (Figure 1 and
Appendix C). In the long run, there is a permanent decrease in the level of real GDP of 2.9
percent. The tariff hike triggers a one-off rise in domestic prices in the first year, pushing up
the Consumer Price Index (CPI) inflation by 0.7 percentage points. Higher inflation prompts
an increase in the nominal interest rate through rule-based monetary policy, creating a
disinflationary effect in subsequent years. Over time, the inflationary impact of the tariff
diminishes, with inflation returning to its target level in the long run.

11. Lower exports and investment drive the contraction of the U.S. real GDP. Higher
tariffs weaken import demand and raise the U.S. domestic prices relative to foreign prices,
leading to a real appreciation of the U.S. dollar. As tariff hikes immediately raise import prices,
U.S. imports decline by 13.8 percent in the initial years. Over time, with the effects of currency
appreciation passing through to import prices, the decline in U.S. imports moderates to 5 to 6
percent. The real appreciation of the U.S. dollar dampens the U.S. exports, which fall by 17
percent in the initial five years and by 20 percent in the long run. The declines in exports,
combined with a stronger U.S. dollar, reduce the profits earned by the U.S. firms from
overseas markets. Moreover, higher costs for imported intermediate goods further erode their
overall profitability. The weaker profit outlook depresses private investment, leading to a
decline of around 5.5 percent in the first five years. As returns on capital gradually recover
alongside slower capital accumulation, the investment contraction moderates over the
medium term, stabilizing at a 3.9 percent decline in the long run. Meanwhile, additional tariff
revenue—estimated at 1.4 percent of U.S. GDP—is partially redistributed to households
through lump-sum transfers. This boosts private consumption in the first decade, partly
offsetting the negative effects of declining exports and investment.

12. China and ROW also experience contractions in export and real GDP. The tariff
hike affects U.S. trade partners through trade and exchange rate linkages. In the first year
following the tariff increase, exports of goods to the U.S. decline by 47.2 percent for China and
11.7 percent for ROW, resulting in an overall export contraction of 7.3 percent and 4.2 percent,
respectively. With the effects of their currency depreciation unfolding, the decline in exports
gradually eases over time. Due to the dollar pricing mechanism in the model, imports of China
and ROW decline sharply in the first year. While their imports from each other progressively
recover, imports from the U.S. remain persistently lower, stabilizing at declines of 25 percent

4
and 22 percent, respectively. Alongside export contraction, both private consumption and
investment decline in China and ROW compared to the baseline. Overall, real GDP contracts
by around 0.5 percent in China and ROW in the first year following the tariff increase. Inflation
in both China and ROW increases as well in the first year following the rises in U.S. tariffs,
mainly driven by the depreciation of their currencies. However, it declines in the second and
third years as the deflationary pressures from weaker growth dominate.

Figure 1. Macroeconomic Impacts under Trump Tariffs Scenario

Real GDP (percent change) Household Consumption (percent change)

0.0 1.0
USA
-0.5 0.5 CHN
ROW
-1.0 0.0
USA
-1.5 CHN -0.5

ROW
-2.0 -1.0

-2.5 -1.5

-3.0 -2.0
0 5 10 15 20 0 5 10 15 20

Investment (percent change) Real Exchange Rate against ths US$* (percent change)

0.0
14.0

-1.0 12.0

-2.0 10.0

-3.0 8.0
CHN ROW
-4.0 6.0

-5.0 4.0
USA

-6.0 CHN 2.0

ROW
-7.0 0.0
0 5 10 15 20 0 5 10 15 20

* + indicates depreciation.

Exports (percent change) Imports (percnet change)

0.0 0.0
-2.0
-2.0
-4.0
-4.0
-6.0

-8.0 -6.0
USA
-10.0 CHN -8.0
USA
-12.0 ROW -10.0 CHN
-14.0 ROW
-12.0
-16.0

-18.0 -14.0

-20.0 -16.0
0 5 10 15 20 0 5 10 15 20

5
 Exports to the U.S. (percent change) Imports from the U.S. (percent change)
0.0 0.0
-5.0 0 5 10 15 20

-10.0 -5.0

-15.0
-10.0
-20.0 CHN CHN
-25.0 -15.0
ROW ROW
-30.0
-35.0 -20.0

-40.0
-25.0
-45.0
-50.0 -30.0
0 5 10 15 20

CPI Inflation (percentage point change) Nominal Interest Rate (percentage point change)
0.8 0.6
USA
0.6 0.5
CHN
USA 0.4
0.4 ROW
CHN
0.3
0.2 ROW
0.2
0.0
0.1

-0.2 0.0

-0.4 -0.1
0 5 10 15 20 0 5 10 15 20

Source: Model simulations

13. In relative terms, the long-term losses in real GDP for China and the ROW are
smaller than those projected for the U.S. The losses in real GDP in China and ROW narrow
slightly after the first year, but gradually widen again, reaching long-term declines of 1.8
percent for China and 1 percent for ROW. Although China faces significantly higher additional
tariffs under the Trump 2.0 measures, the simulation results suggest that its near-term
economic impact is comparable to that of ROW, reflecting China’s lower dependence on trade
with the U.S. This implies that economies with more diversified trade structures may be better
positioned to absorb the negative spillovers from trade disruptions. Although the U.S. is
projected to incur larger relative GDP losses in the long run, the combined economic size of
China and the ROW—approximately three times that of the U.S.—means that the absolute
long-term GDP losses in these economies are greater. The simulation results indicate that for
every one dollar of long-term real GDP loss in the U.S., there is a corresponding loss of
approximately 1.14 dollars in the rest of the world (i.e., China plus ROW).

14. China’s tariff retaliation results in additional GDP losses for both the U.S and
China, while ROW experiences slight benefits. Given that exports to China account for only
0.5 percent of U.S. GDP, it is unsurprising that a 36.5 percent tariff imposed by China on the
U.S. has only a minor impact on the U.S. Compared to the Trump tariffs scenario, U.S. GDP
losses increase by just 0.04 percentage points in the initially years and 0.1 percentage points
in the long run (Figure 2 and Appendix C). However, China’s tariff retaliation has a more
significant negative impact on its own economy, leading to additional output losses of
approximately 0.1 percentage points in the short term and 0.3 percentage points in the long

6
run. With retaliatory tariffs in place, the decline in China’s imports from the U.S. escalates from
23 to 25 percent in the Trump tariffs scenario to 55 to 57 percent. While China’s exports to the
U.S. remain largely unaffected, its exports to ROW increase less, resulting in a larger overall
decline in total exports. ROW gains slightly from China’s retaliation against the U.S., with GDP
losses narrowing by 0.07 percentage points in the long run. This reflects trade diversion effects,
as ROW exporters gain market share in both China and the U.S. at the expense of their
bilateral trade.

15. An escalation in tariff retaliation delivers a significant blow to both the U.S. and
China, exacerbating global economic losses. Under the escalated trade war scenario, U.S.
GDP losses increase from an average of 0.8 percent to 1.1 percent in the first three years and
from 3.0 percent to 4.1 percent in the long run. Meanwhile, the increase in U.S. CPI inflation
in the first year rises from 0.75 to 1.0 percentage points (Figure 2). For China, the impact is
more severe. Its GDP losses increase by roughly 65 percent relative to the China retaliation
scenario, reaching 0.92 percent in the first three years and 3.0 percent in the long run. The
additional retaliatory tariff hikes on both sides triggers a sharp contraction in the U.S.-China
bilateral trade, with Chinese exports to the U.S. plunging by 80 percent in the first year. As the
real depreciation of renminbi (RMB) against the U.S. dollar surges to 12.6 percent, China’s
exports to ROW increase, partially offsetting the losses in the U.S. market. As a result, China’s
total exports shrink by 10.6 percent in the first year. Chinese imports from the U.S. also decline
by 84 percent in the first year, signaling a significant decoupling of the U.S.–China trade
relationship.

Figure 2. Macroeconomic Impacts under the Three Scenarios

Real GDP - U.S. (percent change) Real GDP – China (percent change)

0.0 0.0

-0.5
Trump Tariffs -0.5
-1.0
China Retaliation
-1.5 -1.0
Escalated Trade War
-2.0
-1.5
-2.5

-3.0 -2.0

-3.5 Trump Tariffs

-2.5 China Retaliation
-4.0
Escalated Trade War
-4.5 -3.0
0 5 10 15 20 0 5 10 15 20

Real GDP – Rest of the World (percent change) CPI Inflation – US (percentage point change)
0.0 1.2
-0.1
Trump Tariffs 1.0
-0.2 Trump Tariffs
China Retaliation
-0.3 0.8
Escalated Trade War China Retaliation
-0.4
0.6 Escalated Trade War
-0.5

-0.6 0.4
-0.7
0.2
-0.8

-0.9 0.0
0 5 10 15 20 0 5 10 15 20

7
 China’s Total Exports (percent change) China’s Exports to the U.S. (percent change)

0.0 0.0

-10.0
-2.0 Trump Tariffs
-20.0
China Retaliation
-4.0 -30.0
Escalated Trade War
-40.0
-6.0
-50.0

-8.0 Trump Tariffs -60.0

China Retaliation -70.0

-10.0
Escalated Trade War -80.0

-12.0 -90.0
0 5 10 15 20 0 5 10 15 20

China’s Exports to ROW (percent change) China’s Total Imports (percent chang)

8.0 0.0

7.0 Trump Tariffs

-2.0
6.0 China Retaliation
-4.0
5.0 Escalated Trade War
-6.0
4.0

3.0 -8.0
Trump Tariffs
2.0 -10.0
1.0 China Retaliation
-12.0
0.0
-14.0
-1.0

-2.0 -16.0
0 5 10 15 20 0 5 10 15 20

China’s Imports from the U.S. (percent change) China’s Exchange Rate against the US$* (percent change)

0.0 14.0

-10.0
12.0
-20.0
10.0
-30.0
Trump Tariffs
-40.0 8.0
China Retaliation
-50.0 Escalated Trade War Trump Tariffs
6.0
-60.0 China Retaliation
4.0
-70.0
Escalated Trade War
2.0
-80.0

-90.0 0.0
0 5 10 15 20 0 5 10 15 20
* + indicates depreciation
Source: Model simulations

IV. Possible Policy Responses by China

16. Exchange rate flexibility would play a crucial role in helping China mitigate the
economic shocks, should the U.S. implement higher tariffs under Trump’s trade
policies. The simulation scenarios in the previous section indicate that RMB undergoes a
significant real depreciation following the implementation of these tariffs. A weaker RMB
moderates import demand and enhances export competitiveness, thereby partly offsetting the
adverse impact of higher tariffs on Chinese goods. Because of price rigidities in domestic
goods and factor markets, nominal exchange rate changes are often the primary channel to
achieve real exchange rate adjustment. A less flexible nominal exchange rate system could
hinder this adjustment process, resulting in higher short-term economic costs for China. To

8
assess the implications of reduced exchange rate flexibility, we conduct a separate simulation
for the base-case tariff scenario, assuming an alternative adjustment mechanism where
capital controls prevent full adherence to the interest rate parity condition, causing RMB to
depreciate only partially following the U.S. tariff hikes. Under this alternative assumption, the
real exchange rate of RMB depreciates against the U.S. dollar by just 7.9 percent against the
U.S. dollar in the first year and 10.4 percent in the second year, compared to the 11.8 percent
and 11.6 percent depreciation respectively under the original assumption of more flexible
nominal exchange rate. As a result of the smaller depreciation, China’s real GDP declines by
2.3 percent in the first year and 1.3 percent in the second year, significantly exceeding the
economic losses observed under the original assumption.

Figure 3. Macroeconomic Impacts of the Trump Tariffs under Alternative Exchange

Rate Assumption (percent change from the baseline)
China’s Real GDP Real Exchnage Rate of RMB against the U.S. Dollar

0.0 14.0

12.0
-0.5
10.0

-1.0 8.0
Trump Tariffs Scenario Trump Tariffs Scenario
6.0
-1.5
Trump Traiffs Scenario with Alternative
Trump Traiffs Scenario with Alternative 4.0 Exchange Rate Assumption
Exchange Rate Assumption
-2.0
2.0

0.0
-2.5
0 5 10
0 5 10

Source: Model simulations

17. Fiscal easing is expected to serve as a key policy tool to counter the headwinds
from higher tariffs. In response to the intensified trade tensions under the Trump 2.0 tariffs,
China has ramped up fiscal stimulus to cushion the economic impact and stabilize GDP
growth. During the recent National People's Congress, the Chinese government announced
a series of fiscal stimulus initiatives aimed at supporting the economy amidst external
challenges. China set a higher budget deficit target for 2025 at 4 percent of GDP—the highest
in recent decades. Quotas for both local and central government special bond net issuances
have been raised to fund investment and consumption-supporting initiatives, including
consumer goods trade-in programs, equipment upgrades, and strategic investment projects.
To assess the impact of expansionary fiscal policies, we simulate three fiscal stimulus
scenarios, each assuming a 1 percent of GDP increase in government spending over a two-
year period. The government consumption scenario assumes the additional spending is
allocated to public sector consumption, the general household transfer scenario distributes
the fund as a lump-sum transfer to all households, while the targeted household transfer
scenario directs fiscal transfers only to liquidity-constrained households in the model.

18. A temporary 1 percent-of-GDP increase in government consumption boosts

Chinese GDP by 0.6 percent during the period of fiscal easing. With the fiscal expansion
deployed for two years, real GDP rises during the period of fiscal stimulus, driven by the
increases in both public and private consumption (Figure 4). Increased fiscal expenditure
leads to more employment and higher wages, thereby boosting household consumption.
However, during the stimulus period, China’s investment declines by approximately 1 percent,
reflecting the crowding-out effect caused by rising interest rates. The 0.6 percent increase in

9
real GDP during the period of fiscal expansion suggests a fiscal multiplier of similar magnitude.
After the stimulus, real GDP declines by around 0.3 percent initially, with a long-term GDP
loss remaining around 0.1 percent, as the higher permanent government debt raises interest
rates and reduces capital stock in the long run. This result highlights the fact that fiscal
expansion, while effective as a short-term stimulus, is insufficient to offset the long-term output
and welfare losses resulting from tariffs. 7

Figure 4. Macroeconomic Impacts of China’s Fiscal Stimulus

Real GDP - China (percent change) Ratio of Government Debt to GDP – China (percentage point
change)

0.8 2.5
Government consumption scenario
0.6
General household transfer scenario 2.0
0.4 Targeted household transfer scenario

0.2 1.5

0.0 Government consumption scenario

1.0
General household transfer scenario
-0.2
Targeted household transfer scenario
0.5
-0.4

-0.6 0.0
0 5 10 15 20 0 5 10 15 20

CPI Inflation – China (percentage point change) Nominal Inteest Rate – China (percentage point change)
0.10 0.5

0.4
Government consumption scenario Government consumption scenario
0.05 General household transfer scenario 0.3 General household transfer scenario
Targeted household transfer scenario Targeted household transfer scenario
0.2

0.00 0.1

0.0

-0.05 -0.1
0 5 10 15 20 0 5 10 15 20

Consumption – China (percent change) Invetsment - China (percent change)

3.5 0.0
Government consumption scenario
3.0
General household transfer scenario -0.2
2.5
Targeted household transfer scenario
-0.4
2.0

1.5 -0.6
Government consumption scenario
1.0
-0.8 General household transfer scenario
0.5
Targeted household transfer scenario
-1.0
0.0

-0.5 -1.2
0 5 10 15 20 0 5 10 15 20

Source: Model simulations

7 The model underestimates the positive impact of government spending on economic growth as it assumes such
spending is non-productive and non-utility- generating. Incorporating public investment and public capital in the
model would provide a more accurate assessment of the long-term growth effects of government spending.

10
19. Compared with increasing government consumption, providing direct transfers
to households results in slower economic growth but greater gains in household
consumption. Government transfer directly raises household income, but not all of the
additional income translates into an expansion in aggregate demand. Forward-looking
households—those with access to financial markets—save a portion of their increased income
for future consumption, leading to some leakage in the fiscal stimulus impact. With the model
assumption that 35 percent of households in China are liquidity-constrained and the remaining
65 percent are forward-looking, the effects of government transfers are mixed. When transfers
are distributed across both household types, real GDP rises only marginally by 0.2 percent
over the two years of fiscal easing. Household consumption expands by around 0.7 percent,
as increased income boosts spending, albeit with some savings leakage. By contrast, when
transfers are specifically targeted to lower-income, liquidity-constrained households, who
spend their entire current income, the immediate impact on economic growth is significantly
stronger. In this scenario, real GDP increases by 0.5 percent, while household consumption
surges by 3 percent, demonstrating greater direct benefits to household well-being. As in the
previous scenario, investment declines in both transfer scenarios due to higher interest rates
and associated crowding-out effects.

V. Conclusion

20. The simulation results indicate substantial economic costs associated with a
trade war under Trump 2.0, in which all the countries involved would experience
negative impacts. This note has examined the implications of Trump 2.0 tariffs and possible
policy response by China to mitigate their adverse effects. Specifically, using a calibrated
global dynamic general equilibrium model, we simulate scenarios involving U.S. tariff hikes,
China’s tariff retaliation, and China’s fiscal expansion. Simulation results of these scenarios
suggest that tariff policies adopted by President Trump since his second inauguration would
cause a sharp slowdown in economic growth and higher domestic prices in the U.S.
Meanwhile, China would face output losses and significant declines in trade, but currency
depreciation and fiscal stimulus help alleviate some of these negative impacts. The rest of the
world would also incur important economic losses. Overall, Trump’s protectionist trade agenda
—particularly if it escalates into a full-scale trade war—would impose substantial economic
costs on the U.S., China and the global economy.

21. In response to Trump 2.0 tariffs, China can employ an array of policy measures,
including exchange rate flexibility, accommodative monetary policy, fiscal stimulus,
and structural reform. In the short term, a combination of flexible exchange rate and targeted
fiscal stimulus can cushion the economy from immediate disruptions. Allowing RMB to adjust
freely in a market-driven environment will help sustain export competitiveness amidst turbulent
global trade conditions. While expanding government consumption can lift short-term growth
more effectively, directing fiscal transfers specifically toward lower-income households can
simultaneously achieve economic expansion and promote greater social inclusion. However,
macro stimulus policies cannot resolve the long-term adverse effects associated with Trump
tariffs. Over the medium to long term, China needs to pivot toward strengthening domestic
demand drivers and enhancing industrial capabilities to reduce vulnerability to external
shocks. Accelerating trade diversification, expanding economic ties with emerging markets
and reinforcing supply chain resilience will also be crucial in navigating an increasingly
protectionist global environment. China should reaffirm its commitment to the rules-based
multilateral trading system centered on the World Trade Organization (WTO). Demonstrating
leadership in upholding global trade norms and promoting openness would help safeguard
international trade flows and mitigate the risk of global economic fragmentation.

11
22. Some important caveats of this study are worth noting. First, the model is highly
aggregated and does not capture industry-level heterogeneity and global value chain (GVC)
linkages, which are critical to understanding how tariffs propagate through complex production
networks. Incorporating multi-sector modeling of industrial chains could better reflect the
dynamic nature of global trade structures and capture the tariff-induced reconfiguration of
GVC. Second, the model assumes complete pass-through of tariffs to consumer prices. Future
work could incorporate richer microfoundations for price-setting behavior to enhance the
analysis into the price effects and incidence of tariffs. Third, the current framework is
deterministic and excludes policy uncertainty and economic volatility, thereby likely
underestimating the impacts of an uncertain trade environment on global economy.
Incorporating stochastic shocks would enhance the model’s capacity to reflect real-world
complexities. Fourth, the model assumes constant death rate excludes bequests and
precautionary savings, which may lead to an overestimation of households’ short-term
consumption responses to shocks. Incorporating age-specific mortality rates and
intergenerational transfers would enhance the model’s ability to capture the dynamics of
consumption behavior. Finally, the model does not explicitly incorporate the linkages through
global financial markets and the mechanisms of co-movements in asset prices. As these are
important transmission channels through which a trade war may further weaken the global
economy, our results likely underestimate the impacts of the Trump 2.0 tariffs. These
limitations point to the directions for further research.

12
 References

Anderson, D., B. L. Hunt, M. Kortelainen, M. Kumhof, D. Laxton, D. V. Muir, S. Mursula, and

S. Snudden. 2013. "Getting to Know GIMF: The Simulation Properties of the Global
Integrated Monetary and Fiscal Model," IMF Working Paper Series, 2013/055,
International Monetary Fund

Blanchard, O. 1985. “Debt, Deficits, and Finite Horizons”, Journal of Political Economy, Vol.
93, pp. 223-247.

Cavallo, A., G. Gopinath, B Neiman and J. Tang. 2021. "Tariff Pass-Through at the Border
and at the Store: Evidence from US Trade Policy." American Economic Review:
Insights 3 (1): 19–34

Fajgelbaum, P. and A. K. Khandelwal. 2022. “The Economic Impacts of the US–China Trade
War.” Annual Review of Economics, Vol. 14, pp. 205-228

Goldman Sachs. 2025. “USA: President Trump Announces ‘Reciprocal’ Tariffs.” Economics
Research, April 2, 2025

Gopinath, G. and O. Itskhoki. 2022. “Dominant Currency Paradigm: A Review.” In Handbook

of International Economics, ed. By G. Gopinath, E. Helpman, K. Rogoff, Elsevier,
Volume 6, Chapter 2. Pages 45-90

Kumhof, M. and D. Laxton. 2007. “A Party without a Hangover? On the Effects of U.S. Fiscal
Deficits.” IMF Working Paper Series, 2007/202, International Monetary Fund

Kumhof, M., D. Laxton, D. Muir and S. Mursula. 2010. “The Global Integrated Monetary Fiscal
Model (GIMF) - Theoretical Structure”, IMF Working Paper Series, 2010/34,
International Monetary Fund

Rotemberg, J. 1982. “Sticky Prices in the United States.” Journal of Political Economy
90(6):1187–1211.

Yaari, M. E. 1965. “Uncertain Lifetime, Life Insurance, and the Theory of the Consumer.”
Review of Economic Studies 32(2) 137–150e

13
 Appendix A: Model Specification

(1) Basic Setup

The model economy consists of three countries (regional blocs) which are indexed by r or s.
Each region is populated by overlapping generations households with finite planning horizons
as in Blanchard (1985). Households are indexed by age 𝑎𝑎 . In each region there are a
continuum of firms and a continuum of unions, which are indexed by 𝑛𝑛 ∈[0,1] and 𝑢𝑢 ∈[0,1],
respectively. t is time index in the model.

The model exhibits an exogenous trend in labor productivity (at rate g-1). For a clear
separation of endogenous dynamic from exogenous trend, we present all variables in
detrended form through division by g. In each region the CPI is the numéraire of the economy,
and all national prices are expressed in terms of domestic consumption units.

In the model description below, subscripts denote the set indexes. The time index, t, is omitted
when all variables in an equation are with same time index. Country subscript, r, is also omitted
where doing so does not lead to confusion.

(2) Demand and Trade

Domestic demand of each region comprises consumption and investment made by

households, government consumption and intermediate demand by firms. A composite good
XA is used for final and intermediate demand. The composite good is a constant elasticity of
substitution (CES) aggregation of domestic goods, XD and aggregate imports, XM.
𝑚𝑚 𝑚𝑚 −1)⁄𝜎𝜎 𝑚𝑚 1/𝜎𝜎𝑚𝑚 𝑚𝑚 −1)⁄𝜎𝜎 𝑚𝑚
𝜎𝜎 𝑚𝑚 ⁄(𝜎𝜎𝑚𝑚 −1)
𝑋𝑋𝑋𝑋 = �(𝛼𝛼 𝑚𝑚 )1/𝜎𝜎 (𝑋𝑋𝑋𝑋)(𝜎𝜎 + �𝛼𝛼 𝑑𝑑 � (𝑋𝑋𝑋𝑋)(𝜎𝜎 � (1)

where 𝜎𝜎 𝑚𝑚 is the substitution elasticity between imports and domestic goods. The Armington
share parameters 𝛼𝛼 𝑑𝑑 and 𝛼𝛼 𝑚𝑚 reflect the preference of agents biasing for home or imported
products. The sales price for composite goods, PA, is tax-included dual price index defined
over the prices of imports and domestic goods:
𝑚𝑚 𝑚𝑚 1⁄(1−𝜎𝜎 𝑚𝑚 )
𝑃𝑃𝑃𝑃 = (1 + 𝜏𝜏 𝑠𝑠 )�𝛼𝛼 𝑚𝑚 (𝑃𝑃𝑃𝑃)1−𝜎𝜎 + 𝛼𝛼 𝑚𝑚 (𝑃𝑃𝑃𝑃)1−𝜎𝜎 � (2)
where 𝜏𝜏 𝑠𝑠 is the sales tax rate.

The demand functions generated from (1) and (2) are:

𝑋𝑋𝑋𝑋 𝑃𝑃𝑃𝑃 𝜎𝜎𝑚𝑚

= 𝛼𝛼 𝑑𝑑 � � (3)
𝑋𝑋𝑋𝑋 (1+𝜏𝜏𝑠𝑠 )𝑃𝑃𝑃𝑃
𝑋𝑋𝑋𝑋 𝑃𝑃𝑃𝑃 𝜎𝜎 𝑚𝑚
= 𝛼𝛼 𝑚𝑚 � � (4)
𝑋𝑋𝑋𝑋 (1+𝜏𝜏𝑠𝑠 )𝑃𝑃𝑃𝑃

Aggregate import demand XA is a CES aggregation of imports from each region, i.e.:

𝑤𝑤 𝑤𝑤 −1)⁄𝜎𝜎 𝑤𝑤 𝜎𝜎𝑠𝑠𝑤𝑤 ⁄(𝜎𝜎𝑠𝑠𝑤𝑤 −1)

𝑋𝑋𝑀𝑀𝑠𝑠 = �∑𝑟𝑟∈𝑅𝑅(𝛼𝛼𝑟𝑟𝑟𝑟 )1/𝜎𝜎𝑠𝑠 (𝑊𝑊𝑊𝑊𝐹𝐹𝑟𝑟𝑟𝑟 )(𝜎𝜎𝑠𝑠 𝑠𝑠 � (5)

where WTFrs represents the quantity of good produced in region r sold in the market of region
s and 𝜎𝜎 𝑤𝑤 is the second-level Armington elasticity of substitution among imports from different
regions. The dual price index of aggregate import, PMs, is defined over the prices of each
import supplier, PErs:

14
 𝑤𝑤
𝜀𝜀𝑠𝑠 1−𝜎𝜎𝑠𝑠𝑤𝑤 1⁄(1−𝜎𝜎𝑠𝑠 )
𝑃𝑃𝑀𝑀𝑠𝑠 = �∑𝑟𝑟∈𝑅𝑅 𝛼𝛼𝑟𝑟𝑟𝑟 �𝜀𝜀 (1 + 𝜏𝜏𝑟𝑟𝑟𝑟 )𝑃𝑃𝐸𝐸𝑟𝑟𝑟𝑟 � � (6)
𝑟𝑟

where 𝜏𝜏𝑟𝑟𝑟𝑟 is import tariff rate and 𝜀𝜀𝑟𝑟 is the CPI-based real exchange rate, expressed as the
price of one unit US consumption in terms of domestic consumption.

The demand function generated from (5) and (6) is:

𝑊𝑊𝑊𝑊𝐹𝐹𝑟𝑟𝑟𝑟 𝜀𝜀 𝑃𝑃𝑀𝑀𝑠𝑠 𝜎𝜎𝑠𝑠𝑤𝑤
= 𝛼𝛼𝑟𝑟𝑟𝑟 �𝜀𝜀𝑟𝑟 (1+𝜏𝜏 � (7)
𝑋𝑋𝑀𝑀𝑠𝑠 𝑠𝑠 𝑟𝑟𝑟𝑟 )𝑃𝑃𝐸𝐸𝑟𝑟𝑟𝑟

Each firm is assumed to produce differentiated product, and each variety is an equally
imperfect substitute for all others across all varieties. The aggregate demand for domestic
goods, XD, and aggregate exports, WTF, are further decomposed into demand for variety
provided by each firm, following the standard Dixit-Stiglitz framework:
𝑓𝑓
𝑛𝑛 𝜎𝜎𝑠𝑠
𝑥𝑥𝑑𝑑𝑛𝑛,𝑠𝑠 𝑃𝑃𝐷𝐷𝑠𝑠
= �𝑝𝑝 � (8)
𝑋𝑋𝐷𝐷𝑠𝑠 𝑛𝑛,𝑠𝑠
𝑓𝑓
𝑛𝑛 𝜎𝜎𝑠𝑠
𝑤𝑤𝑤𝑤𝑓𝑓𝑛𝑛,𝑟𝑟𝑟𝑟 𝑃𝑃𝐸𝐸𝑟𝑟𝑟𝑟
= � 𝑝𝑝 � (9)
𝑊𝑊𝑊𝑊𝐹𝐹𝑟𝑟𝑟𝑟 𝑛𝑛,𝑟𝑟

where 𝜎𝜎 𝑓𝑓 is the substitution elasticity among varieties of each firm and pn,s represents the price
𝑛𝑛 𝑛𝑛
of variety n in region s. 𝑥𝑥𝑑𝑑𝑛𝑛,𝑠𝑠 and 𝑤𝑤𝑤𝑤𝑓𝑓𝑛𝑛,𝑟𝑟𝑟𝑟 represent the quantity of domestic demands for
variety produced by firm n in region s, and the demand for variety produced by firm n in region
r and exported to region s, respectively.

(3) Firms

Production technology of firms is modeled using nested CES functions. At the top level, the
output is split into intermediate input and a bundle of capital and labor input. At the second
level, the bundle of capital and labor inputs is disaggregated into capital and aggregate labor.
Finally, at the bottom level, aggregate labor is decomposed into the differentiated labor input
by each union.

Each firm produces a different variety and sets the price of its products in face of isoelastic
demand functions in both domestic and foreign markets, as shown in (8) and (9). There is
adjustment cost for price setting, which, expressed as a proportion of total sales, is assumed
to be given by the following functions:

𝑝𝑝 2
𝑝𝑝𝑝𝑝 𝜑𝜑𝑟𝑟 𝑝𝑝𝑑𝑑𝑛𝑛,𝑟𝑟,𝑡𝑡 /𝑝𝑝𝑑𝑑𝑛𝑛,𝑟𝑟,𝑡𝑡−1
𝛤𝛤𝑛𝑛,𝑟𝑟,𝑡𝑡 = �𝜋𝜋𝑟𝑟,𝑡𝑡 − 1� (10)
2 𝛱𝛱𝑟𝑟,𝑡𝑡
𝑝𝑝 2
𝑝𝑝𝑝𝑝 𝜑𝜑𝑠𝑠 𝜀𝜀𝑡𝑡−1,𝑟𝑟 𝑝𝑝𝑒𝑒𝑛𝑛,𝑟𝑟,𝑠𝑠,𝑡𝑡 /𝑝𝑝𝑒𝑒𝑛𝑛,𝑟𝑟,𝑠𝑠,𝑡𝑡−1
𝛤𝛤𝑛𝑛,𝑟𝑟,𝑠𝑠,𝑡𝑡 = �𝜋𝜋𝑟𝑟,𝑡𝑡 − 1� (11)
2 𝜀𝜀𝑡𝑡,𝑟𝑟 𝛱𝛱𝑠𝑠,𝑡𝑡

where π is domestic CPI inflation rate. The price adjustment cost functions indicate that the
adjustment cost is related to changes of nominal prices of products relative to the
contemporaneous inflation target for the CPI, indexed by П. (11) indicates that export prices
are set in U.S. dollar, i.e. dominant currency pricing.

A firm n is assumed to maximize the discounted value of current and future dividends, divn:

15
 𝑔𝑔𝜏𝜏 𝜋𝜋𝜏𝜏,𝑟𝑟
𝑚𝑚𝑚𝑚𝑚𝑚 ∑∞ 𝑡𝑡
𝑡𝑡=0 ∏𝜏𝜏=0 𝑑𝑑𝑑𝑑𝑣𝑣𝑛𝑛,𝑟𝑟,𝑡𝑡
1+𝑖𝑖𝜏𝜏,𝑟𝑟
𝑔𝑔 𝑔𝑔
𝑛𝑛
𝑑𝑑𝑑𝑑𝑣𝑣𝑛𝑛,𝑟𝑟,𝑡𝑡 = �1 − 𝜏𝜏𝑟𝑟𝑘𝑘 �𝑅𝑅𝑟𝑟,𝑡𝑡 𝐾𝐾𝑛𝑛,𝑟𝑟,𝑡𝑡 − 𝐼𝐼𝑛𝑛,𝑟𝑟,𝑡𝑡 + 𝑅𝑅𝑟𝑟,𝑡𝑡 𝐾𝐾𝑛𝑛,𝑟𝑟,𝑡𝑡 (12)
𝑝𝑝𝑝𝑝 𝑝𝑝𝑝𝑝
+(𝑝𝑝𝑑𝑑𝑛𝑛,𝑟𝑟,𝑡𝑡 − 𝑃𝑃𝑋𝑋𝑟𝑟,𝑡𝑡 )𝑥𝑥𝑑𝑑𝑛𝑛,𝑟𝑟,𝑡𝑡 (1 − 𝛤𝛤𝑛𝑛,𝑟𝑟,𝑡𝑡 ) + ∑𝑠𝑠∈𝑅𝑅(𝑝𝑝𝑒𝑒𝑛𝑛,𝑟𝑟,𝑠𝑠,𝑡𝑡 − 𝑃𝑃𝑋𝑋𝑟𝑟,𝑡𝑡 )𝑤𝑤𝑤𝑤𝑓𝑓𝑛𝑛,𝑟𝑟,𝑠𝑠,𝑡𝑡 (1 − 𝛤𝛤𝑛𝑛,𝑟𝑟,𝑠𝑠,𝑡𝑡 )

subject to the CES production technology, the demand functions of (8) and (9), the adjustment
costs in price setting of (10) and (11), given marginal cost PX, and the law of motion of capital:
𝐼𝐼
𝐾𝐾𝑛𝑛,𝑡𝑡+1 𝑔𝑔𝑡𝑡+1 = (1 − 𝛿𝛿)𝐾𝐾𝑛𝑛,𝑡𝑡 + 𝛤𝛤𝑛𝑛,𝑖𝑖,𝑡𝑡 𝐾𝐾𝑛𝑛,𝑡𝑡 (13)

where δ is the depreciation rate. Γ𝑛𝑛𝐼𝐼 is a function of investment/capital ratio with adjustment
costs are zero in the steady state:
2
𝐼𝐼 𝐼𝐼 𝜑𝜑 𝐼𝐼 𝐼𝐼𝑛𝑛,𝑡𝑡 𝐼𝐼
𝛤𝛤𝑛𝑛,𝑡𝑡 = 𝐾𝐾𝑛𝑛,𝑡𝑡 − � − 𝐾𝐾𝑡𝑡−1 � (14)
𝑛𝑛,𝑡𝑡 2 𝐾𝐾𝑛𝑛,𝑡𝑡 𝑡𝑡−1

As shown in (12), the firm’s dividends include the after-tax return to its private capital, the
return to public capital captured by individual firm, and the profits the firm obtains from selling
in domestic market and exports. The optimization problem of firm is to determine its levels of
investment, labor and intermediate inputs, and set the nominal prices of its products in
domestic and export markets to maximize the discounted present value of its dividends. The
resulting first order conditions with respect to I and K are:
1 𝐼𝐼 𝐼𝐼
𝑞𝑞𝑡𝑡
= 1 − 𝜑𝜑 𝐼𝐼 �𝐾𝐾𝑛𝑛,𝑡𝑡 − 𝐾𝐾𝑡𝑡−1 � (15)
𝑛𝑛,𝑡𝑡 𝑡𝑡−1
𝐼𝐼 1+𝑖𝑖𝑡𝑡
(1 − 𝜏𝜏𝑡𝑡𝑘𝑘 )𝑅𝑅𝑡𝑡 + 𝑞𝑞𝑡𝑡 �1 − 𝛿𝛿 + 𝛤𝛤𝑛𝑛,𝑡𝑡
𝐼𝐼
� − 𝐾𝐾𝑡𝑡 − 𝑞𝑞𝑡𝑡−1 =0 (16)
𝑡𝑡 𝜋𝜋𝑡𝑡

where q is the shadow price of capital, i.e. the Tobin’s Q.

The resulting first order conditions with respect to pd and pe are:

𝑝𝑝𝑝𝑝 𝑝𝑝𝑝𝑝
𝑝𝑝𝑝𝑝 𝑓𝑓 𝑓𝑓 𝑝𝑝 𝜋𝜋 𝜋𝜋𝑛𝑛,𝑟𝑟,𝑡𝑡
�1 − 𝛤𝛤𝑛𝑛,𝑟𝑟,𝑡𝑡 ��𝑝𝑝𝑑𝑑𝑛𝑛,𝑟𝑟,𝑡𝑡 (1 − 𝜎𝜎𝑟𝑟 ) + 𝑃𝑃𝑋𝑋𝑟𝑟,𝑡𝑡 𝜎𝜎𝑟𝑟 � = (𝑝𝑝𝑑𝑑𝑛𝑛,𝑟𝑟,𝑡𝑡 − 𝑃𝑃𝑋𝑋𝑟𝑟,𝑡𝑡 )𝜑𝜑𝑟𝑟 � 𝛱𝛱𝑛𝑛,𝑟𝑟,𝑡𝑡 − 1� 𝛱𝛱𝑟𝑟,𝑡𝑡
𝑟𝑟,𝑡𝑡
𝑝𝑝𝑝𝑝 𝑝𝑝𝑝𝑝
(17)
𝑔𝑔𝑡𝑡+1 𝜋𝜋𝑡𝑡+1 𝑋𝑋𝐷𝐷𝑟𝑟,𝑡𝑡+1 𝑝𝑝 𝜋𝜋𝑛𝑛,𝑟𝑟,𝑡𝑡+1 𝜋𝜋𝑛𝑛,𝑟𝑟,𝑡𝑡+1
− 1+𝑖𝑖 𝑋𝑋𝐷𝐷𝑟𝑟,𝑡𝑡
(𝑝𝑝𝑑𝑑𝑛𝑛,𝑟𝑟,𝑡𝑡+1 − 𝑃𝑃𝑋𝑋𝑟𝑟,𝑡𝑡+1 )𝜑𝜑𝑟𝑟 � 𝛱𝛱 − 1� 𝛱𝛱𝑟𝑟,𝑡𝑡+1
𝑡𝑡+1 𝑟𝑟,𝑡𝑡+1
𝑝𝑝𝑝𝑝 𝑝𝑝𝑝𝑝
𝑝𝑝𝑝𝑝 𝑓𝑓 𝑓𝑓 𝑝𝑝 𝜋𝜋𝑛𝑛,𝑟𝑟,𝑠𝑠,𝑡𝑡 𝜋𝜋𝑛𝑛,𝑟𝑟,𝑠𝑠,𝑡𝑡
�1 − 𝛤𝛤𝑛𝑛,𝑟𝑟,𝑠𝑠,𝑡𝑡 ��𝑝𝑝𝑒𝑒𝑛𝑛,𝑟𝑟,𝑠𝑠,𝑡𝑡 (1 − 𝜎𝜎𝑠𝑠 ) + 𝑃𝑃𝑋𝑋𝑟𝑟,𝑡𝑡 𝜎𝜎𝑠𝑠 � = (𝑝𝑝𝑒𝑒𝑛𝑛,𝑟𝑟,𝑠𝑠,𝑡𝑡 − 𝑃𝑃𝑋𝑋𝑟𝑟,𝑡𝑡 )𝜑𝜑𝑠𝑠 � 𝛱𝛱𝑠𝑠,𝑡𝑡
− 1�
𝛱𝛱𝑠𝑠,𝑡𝑡
𝑝𝑝𝑝𝑝 𝑝𝑝𝑝𝑝 (18)
𝑔𝑔 𝜋𝜋 𝑊𝑊𝑊𝑊𝐹𝐹𝑟𝑟,𝑠𝑠,𝑡𝑡+1 𝑝𝑝 𝜋𝜋 𝜋𝜋𝑛𝑛,𝑟𝑟,𝑠𝑠.𝑡𝑡+1
− 𝑡𝑡+1 𝑡𝑡+1 (𝑝𝑝𝑒𝑒𝑛𝑛,𝑟𝑟,𝑠𝑠,𝑡𝑡+1 − 𝑃𝑃𝑋𝑋𝑟𝑟,𝑡𝑡+1 )𝜑𝜑𝑠𝑠 � 𝑛𝑛,𝑟𝑟,𝑠𝑠.𝑡𝑡+1 − 1�
1+𝑖𝑖𝑡𝑡+1 𝑊𝑊𝑊𝑊𝐹𝐹𝑟𝑟,𝑠𝑠,𝑡𝑡 𝛱𝛱𝑠𝑠,𝑡𝑡+1 𝛱𝛱𝑠𝑠,𝑡𝑡+1

𝑝𝑝𝑝𝑝 𝑝𝑝𝑝𝑝
where 𝜋𝜋𝑛𝑛,𝑟𝑟,𝑡𝑡 is the inflation rate of variety n in domestic market and 𝜋𝜋𝑛𝑛,𝑟𝑟,𝑠𝑠,𝑡𝑡 is the inflation rate
of variety n produced in country r and sold in country s.

The first order conditions with respect to production inputs lead to the following demand
functions and price indices of aggregate inputs:
𝑉𝑉𝑉𝑉 = 𝛼𝛼 𝑣𝑣 ⋅ 𝑋𝑋𝑋𝑋 (19)
𝑋𝑋𝑋𝑋𝑋𝑋 = 𝛼𝛼 𝑛𝑛 ⋅ 𝑋𝑋𝑋𝑋 (20)
𝑣𝑣 𝑛𝑛
𝑃𝑃𝑃𝑃 = 𝛼𝛼 𝑃𝑃𝑃𝑃𝑃𝑃 + 𝛼𝛼 𝑃𝑃𝑃𝑃 (21)
𝑃𝑃𝑃𝑃𝑃𝑃 𝜎𝜎𝑣𝑣
𝐿𝐿 = 𝛼𝛼 𝑙𝑙 � � 𝑉𝑉𝑉𝑉 (22)
𝑊𝑊

16
 𝑣𝑣
𝑃𝑃𝑃𝑃𝑃𝑃 𝜎𝜎
𝐾𝐾 = 𝛼𝛼 𝑘𝑘 � � 𝑉𝑉𝑉𝑉 (23)
𝑅𝑅
𝑣𝑣 𝑣𝑣 1/(1−𝜎𝜎 𝑣𝑣 )
𝑃𝑃𝑃𝑃𝑃𝑃 = �𝛼𝛼 𝑙𝑙 (𝑊𝑊 )1−𝜎𝜎 + 𝛼𝛼 𝑘𝑘 (𝑅𝑅)1−𝜎𝜎 � (24)

where XP, VA, XAP, L, K represent output, aggregate primary factor, intermediate input,
aggregate labor and capital, respectively, and PX, PVA, PA, W and R are their corresponding
price indices. σv is elasticity of substitution between labor and capital.

Firms have the CES aggregator of the differentiated labor varieties provided by unions. As
firms are assumed to be identical, the aggregate labor demand, L, can be expressed as:
𝜎𝜎𝑙𝑙
𝜎𝜎𝑙𝑙−1 𝜎𝜎𝑙𝑙 −1
1
𝐿𝐿 = �∫0 (𝑙𝑙𝑢𝑢 ) 𝜎𝜎𝑙𝑙 𝑑𝑑𝑑𝑑� (25)

where 𝑙𝑙𝑢𝑢 is the labor provided by union u and 𝜎𝜎 𝑙𝑙 is the elasticity of substitution across labor
varieties. Cost minimization of firms implies that demand for labor u is a function of the relative
wage:
𝑙𝑙𝑢𝑢 𝑤𝑤 −𝜎𝜎𝑙𝑙
= � 𝑊𝑊𝑢𝑢 � (26)
𝐿𝐿

where wu is the wage paid to labor u and the wage index W is defined as:
1
1 𝑙𝑙 𝑙𝑙
𝑊𝑊 = �∫0 (𝑤𝑤𝑢𝑢 )1−𝜎𝜎 𝑑𝑑𝑑𝑑�1−𝜎𝜎 (27)

(4) Households

In each period, mr(1-θ) individuals are born in country r and they face a constant probability of
death (1-θ) after their birth. This implies that the total population in country r is mr. We
distinguish between two types of households: forward-looking ones denoted by FL, and
liquidity-constrained ones denoted by LC. For a representative household of generation a, its
period utility in time t, ua,t, is a function of its (detrended) consumption c and labor effort lh.

ℎ 1 ℎ 𝜂𝜂 1−𝜂𝜂 1−𝜎𝜎
𝑢𝑢𝑎𝑎,𝑡𝑡 (𝑐𝑐𝑎𝑎,𝑡𝑡 , 𝑙𝑙𝑎𝑎,𝑡𝑡 ) = 1−𝜎𝜎 ��𝑐𝑐𝑎𝑎,𝑡𝑡 ⁄(𝑐𝑐̃𝑡𝑡−1 )𝑣𝑣 � �1 − 𝑙𝑙𝑎𝑎,𝑡𝑡 � � (28)

where 𝜎𝜎 is the inverse elasticity of intertemporal substitution and 𝜂𝜂 is the weight of

consumption in utility function. The term 𝑐𝑐̃𝑡𝑡−1 represents past per capita consumption of
household h’s peers, i.e. FL households or LC households. v parameterizes the degree of
habit persistence. This is the “catching up with the Joneses” type of external habit formation.
The lifetime utility of age a household at time t, Ua,t, is the sum of discounted period utility:

𝑈𝑈𝑎𝑎,𝑡𝑡 = ∑∞
𝜏𝜏=0(𝑔𝑔
1−𝜎𝜎
𝛽𝛽𝛽𝛽)𝜏𝜏 𝑢𝑢𝑎𝑎+𝜏𝜏,𝑡𝑡+𝜏𝜏 (29)
where 𝛽𝛽 is the subjective discount rate, possibly time-variable but converging to a steady state
constant in the long run.

The decision problem of forward-looking household is to maximize its lifetime utility (29)
subject to following sequences of period budget constraints:

17
 ∗ ∗
𝜃𝜃�𝐵𝐵𝑎𝑎+1,𝑡𝑡+1 𝜋𝜋𝑡𝑡+1 + 𝜀𝜀𝑡𝑡 𝐵𝐵𝑎𝑎+1,𝑡𝑡+1 𝜋𝜋𝑡𝑡+1 + 𝑉𝑉𝑡𝑡+1 𝑥𝑥𝑎𝑎,𝑡𝑡+1 𝜋𝜋𝑡𝑡+1 �𝑔𝑔𝑡𝑡+1 = (1 + 𝑖𝑖𝑡𝑡 )𝐵𝐵𝑎𝑎,𝑡𝑡
∗ ∗
+(1 + 𝑖𝑖𝑡𝑡 )(1 − 𝜁𝜁𝑡𝑡 )𝜀𝜀𝑡𝑡 𝐵𝐵𝑎𝑎,𝑡𝑡 + (𝑉𝑉𝑡𝑡+1 𝜋𝜋𝑡𝑡+1 𝑔𝑔𝑡𝑡+1 𝜃𝜃 + 𝐷𝐷𝐷𝐷𝑣𝑣𝑡𝑡 )𝑥𝑥𝑎𝑎,𝑡𝑡 (30)
+𝑇𝑇𝑅𝑅𝑎𝑎𝐹𝐹𝐹𝐹 𝑡𝑡 + (1 − 𝜏𝜏𝑙𝑙 )𝑤𝑤𝑡𝑡ℎ 𝜑𝜑𝑎𝑎 𝑙𝑙𝑎𝑎,𝑡𝑡
ℎ
− 𝑐𝑐𝑎𝑎,𝑡𝑡 − 𝑇𝑇𝑇𝑇𝑎𝑎,𝑡𝑡

In the above expression, Ba is holdings of domestic government bonds by the representative

household at age a, denominated in domestic currency. 𝐵𝐵𝑎𝑎∗ is holdings of international bonds
by the representative household at age a, denominated in U.S. dollar. π and π* are domestic
CPI inflation rate and the U.S. inflation rate, respectively. i is domestic nominal interest rate
and i* is the nominal interest rate in the U.S. 𝜁𝜁 is the risk premium on the international bonds.
V denotes the value of a claim to profits of firms in current and all future periods and 𝑥𝑥𝑎𝑎 is the
share of firms owned by the representative household at age a. Div is the total dividends paid
by all firms to households. TRFl represents revenue from unions’ profits rebated to forward-
looking households in a lump-sum way. TT is the lump-sum net taxes for households. Labor
incomes 𝑤𝑤 ℎ 𝜑𝜑𝑎𝑎 𝑙𝑙𝑎𝑎ℎ are taxed at the rate 𝜏𝜏 𝑙𝑙 . 𝜑𝜑𝑎𝑎 is the labor productivity of age group a, given by:

𝑚𝑚−𝜃𝜃𝜃𝜃
𝜑𝜑𝑎𝑎 = 𝑚𝑚−𝜃𝜃
𝜒𝜒 𝑎𝑎 (31)

where χ<1 determines the speed of productivity decline of an individual household’s labor
throughout his lifetime.

The first order conditions of the forward-looking household’s optimization problem with respect
to B, B*, C, l and x yield to following arbitrage equations:
𝜋𝜋𝑡𝑡+1 𝜋𝜋∗ 𝜀𝜀𝑡𝑡
1+𝑖𝑖𝑡𝑡+1
= (1+𝑖𝑖∗ 𝑡𝑡+1
)(1−𝜁𝜁) 𝜀𝜀
(32)
𝑡𝑡+1 𝑡𝑡+1
(1−𝜂𝜂)(1−1/𝛾𝛾)
𝑔𝑔𝑡𝑡+1 𝐶𝐶𝑎𝑎,𝑡𝑡+1 (1+𝑖𝑖𝑡𝑡 )𝛽𝛽𝑡𝑡+1 1/𝛾𝛾 𝑤𝑤𝑡𝑡+1
ℎ
𝐶𝐶𝑡𝑡 𝑣𝑣𝑣𝑣(1−1/𝛾𝛾)
𝐽𝐽𝑡𝑡+1 = 𝐶𝐶𝑎𝑎,𝑡𝑡
=� 𝜋𝜋𝑡𝑡+1
� � ℎ
𝑤𝑤𝑡𝑡
𝜒𝜒� �𝐶𝐶𝑡𝑡−1
𝑔𝑔𝑡𝑡+1 � (33)
𝐶𝐶𝑎𝑎,𝑡𝑡 𝜂𝜂𝐹𝐹𝐹𝐹
ℎ
1-𝑙𝑙𝑎𝑎,𝑡𝑡
= 1−𝜂𝜂𝐹𝐹𝐹𝐹 𝑤𝑤𝑡𝑡ℎ 𝜑𝜑𝑎𝑎 (34)

𝑉𝑉𝑡𝑡 (1 + 𝑖𝑖𝑡𝑡 ) = 𝐷𝐷𝐷𝐷𝑣𝑣𝑡𝑡 + 𝑉𝑉𝑡𝑡+1 𝑔𝑔𝑡𝑡+1 𝜋𝜋𝑡𝑡+1 (35)

With some algebraic derivations, the aggregate consumption of all forward-looking

households can be expressed as a fraction of the sum of their financial wealth, FW, and human
wealth, HW:
𝐶𝐶𝑡𝑡𝐹𝐹𝐹𝐹 𝛩𝛩𝑡𝑡 = (1 + 𝑖𝑖𝑡𝑡 )(𝐻𝐻𝑊𝑊𝑡𝑡 + 𝐹𝐹𝑊𝑊𝑡𝑡 ) (36)
𝐹𝐹𝑊𝑊𝑡𝑡 = 𝐵𝐵𝑡𝑡 + 𝜀𝜀𝑡𝑡-1 𝐵𝐵𝑡𝑡∗ + 𝑉𝑉𝑡𝑡 (37)
𝐻𝐻𝑊𝑊𝑡𝑡 = 𝐻𝐻𝐻𝐻𝐿𝐿𝑡𝑡 + 𝐻𝐻𝐻𝐻𝐾𝐾𝑡𝑡 (38)
𝐻𝐻𝐻𝐻𝐿𝐿𝑡𝑡 (1 + 𝑖𝑖𝑖𝑖 ) = 𝑤𝑤𝑡𝑡ℎ 𝐿𝐿𝐹𝐹𝐹𝐹
𝑡𝑡 + 𝐻𝐻𝐻𝐻𝐿𝐿𝑡𝑡+1 𝑔𝑔𝑡𝑡+1 𝜋𝜋𝑡𝑡+1 𝜃𝜃𝜃𝜃 (39)
𝐻𝐻𝐻𝐻𝐾𝐾𝑡𝑡 (1 + 𝑖𝑖𝑖𝑖 ) = 𝑇𝑇𝑅𝑅𝑡𝑡𝐹𝐹𝐹𝐹 + 𝐻𝐻𝐻𝐻𝐾𝐾𝑡𝑡+1 𝑔𝑔𝑡𝑡+1 𝜋𝜋𝑡𝑡+1 𝜃𝜃 (40)

𝛩𝛩−1 is the marginal propensity to consume out of total wealth. The inverse of the marginal
propensity of consumption evolves according to:
1 𝜃𝜃𝐽𝐽𝑡𝑡+1 𝜋𝜋𝑡𝑡+1
𝛩𝛩𝑡𝑡 = + 𝛩𝛩𝑡𝑡+1 (41)
𝜂𝜂𝐹𝐹𝐹𝐹 1+𝑖𝑖𝑡𝑡+1

where J is defined in (33).

Liquidity-constrained households have no access to capital markets. Their decision problem

is purely static, confined to the choices of labor supply. Their budget constraints are:
𝑐𝑐𝑎𝑎,𝑡𝑡 = (1 − 𝜏𝜏 𝑙𝑙 )𝑤𝑤𝑡𝑡ℎ 𝜑𝜑𝑎𝑎 𝑙𝑙𝑎𝑎,𝑡𝑡
ℎ 𝐿𝐿𝐿𝐿
+ 𝑇𝑇𝑇𝑇𝑎𝑎,𝑡𝑡 − 𝑇𝑇𝑇𝑇𝑎𝑎,𝑡𝑡 (42)

18
The first order conditions with respect to consumption and labor supply leads to following
relationship between aggregate consumption and labor supply:
𝐶𝐶𝑡𝑡𝐿𝐿𝐿𝐿 𝜂𝜂𝐿𝐿𝐿𝐿
𝐿𝐿𝐿𝐿 = 𝑤𝑤𝑡𝑡ℎ (43)
𝑚𝑚⋅𝑠𝑠 LC −𝐿𝐿𝑡𝑡 1−𝜂𝜂𝐿𝐿𝐿𝐿

where sLC is the share of liquidity-constrained agents in total households and LLC is the effective
aggregate labor supply of liquidity-constrained households.

(5) Unions

In each there is a continuum of unions which buy labor from households and sell labor to firms.
They are perfectly competitive in their input markets and monopolistically competitive in their
output market. Each union has power to set the nominal wage of the labor they provide.
Similar to the price setting by firms, wage changes are subject to adjustment costs. The
adjustment costs function of nominal wage is assumed as follows:
𝜑𝜑 𝑤𝑤 𝑤𝑤𝑢𝑢,𝑡𝑡 /𝑤𝑤𝑢𝑢,𝑡𝑡−1 2
𝑤𝑤
𝛤𝛤𝑢𝑢,𝑡𝑡 = �𝜋𝜋𝑡𝑡 − 1� (44)
2 𝛱𝛱𝑡𝑡

The decision problem of union is to maximize the present discounted value of nominal wages
paid by firms minus nominal wages paid out to households, minus nominal wage inflation
adjustment costs, by setting the nominal wage:
𝑔𝑔𝜏𝜏 𝜋𝜋𝜏𝜏,𝑟𝑟
𝑚𝑚𝑚𝑚𝑚𝑚 ∑∞ 𝑡𝑡
𝑡𝑡=0 ∏𝜏𝜏=0 1+𝑖𝑖𝜏𝜏,𝑟𝑟
�𝑤𝑤𝑢𝑢,𝜏𝜏 (1-Γ𝑤𝑤 ℎ
u,𝜏𝜏 ) − 𝑤𝑤𝜏𝜏 �𝑙𝑙𝑢𝑢,𝜏𝜏 (45)

subject to demand function (29). The resulting wage setting equation is:
𝑤𝑤
𝑤𝑤𝑡𝑡ℎ 𝜎𝜎𝑙𝑙 𝑤𝑤 𝜋𝜋 𝑤𝑤 𝜋𝜋ℎ,𝑡𝑡
= (𝜎𝜎 𝑙𝑙 − 1)(1 − 𝛤𝛤𝑢𝑢,𝑡𝑡 ) + 𝜑𝜑 𝑤𝑤 � 𝛱𝛱𝑢𝑢,𝑡𝑡 − 1�
(1−𝜏𝜏𝑤𝑤 )𝑤𝑤𝑢𝑢,𝑡𝑡 𝑡𝑡 𝛱𝛱𝑡𝑡
𝑤𝑤 𝑤𝑤 2 (46)
𝑔𝑔𝑡𝑡+1 𝐿𝐿𝑡𝑡+1 𝑤𝑤 𝜋𝜋ℎ,𝑡𝑡+1 �𝜋𝜋ℎ,𝑡𝑡+1 �
− 1+𝑖𝑖 𝜑𝜑 � 𝛱𝛱 − 1�
𝑡𝑡+1 𝐿𝐿𝑡𝑡 𝑡𝑡+1 𝛱𝛱𝑡𝑡+1

𝑤𝑤
where 𝜋𝜋𝑢𝑢,𝑡𝑡 = 𝜋𝜋𝑡𝑡 ⋅ 𝑤𝑤𝑢𝑢,𝑡𝑡 /𝑤𝑤𝑢𝑢,𝑡𝑡−1 is the wage inflation rate.

(6) Government

The government has the following budget constraint:

𝑐𝑐
𝐵𝐵𝑠𝑠,𝑡𝑡+1 𝑔𝑔𝑡𝑡+1 𝜋𝜋𝑡𝑡+1 = (1 + 𝑖𝑖𝑠𝑠,𝑡𝑡 )𝐵𝐵𝑠𝑠,𝑡𝑡 + 𝐺𝐺𝑠𝑠,𝑡𝑡 − 𝜏𝜏𝑠𝑠𝑘𝑘 𝑅𝑅𝑠𝑠,𝑡𝑡 𝐾𝐾𝑠𝑠,𝑡𝑡 − 𝜏𝜏𝑠𝑠𝑙𝑙 𝑊𝑊𝑠𝑠,𝑡𝑡 𝐿𝐿𝑠𝑠,𝑡𝑡 − 𝑇𝑇𝑇𝑇𝑠𝑠,𝑡𝑡 𝑚𝑚𝑠𝑠
(47)
− 𝜏𝜏𝑠𝑠𝑠𝑠 𝑃𝑃𝐴𝐴𝑠𝑠,𝑡𝑡 𝑋𝑋𝐴𝐴𝑠𝑠,𝑡𝑡 − ∑𝑟𝑟�𝜏𝜏𝑟𝑟𝑟𝑟 𝑃𝑃𝐸𝐸𝑟𝑟𝑟𝑟,𝑡𝑡 𝑊𝑊𝑊𝑊𝐹𝐹𝑟𝑟𝑟𝑟,𝑡𝑡 𝜀𝜀𝑠𝑠,𝑡𝑡 ⁄𝜀𝜀𝑟𝑟,𝑡𝑡 �

The central bank in each region is assumed to provide a nominal anchor by employing the
following monetary policy rule:
𝑇𝑇
𝑖𝑖𝑡𝑡 = 𝑖𝑖𝑡𝑡−1 + 𝜔𝜔𝜋𝜋 (𝜋𝜋𝑡𝑡−1 − 𝛱𝛱𝑡𝑡−1 ) + 𝜔𝜔𝑦𝑦 (𝛥𝛥𝛥𝛥𝛥𝛥𝑃𝑃𝑡𝑡−1 − 𝛥𝛥𝛥𝛥𝛥𝛥𝑃𝑃𝑡𝑡−1 ) (48)

where 𝜔𝜔𝜋𝜋 and 𝜔𝜔𝑦𝑦 are weight parameters for inflation gap and output growth gap, respectively.

(7) Equilibrium

Equilibrium in the composite good market. The intermediate inputs, household consumption,
and other final demands constitute the total demand for the same Armington composite goods.

19
 𝑋𝑋𝑋𝑋 = 𝐶𝐶 𝐹𝐹𝐹𝐹 + 𝐶𝐶 𝐿𝐿𝐿𝐿 + 𝐺𝐺 𝐶𝐶 + 𝐼𝐼 (49)

Equilibrium in the factor markets. Labor market clearing conditions in each region can be
specified as follows:
𝐿𝐿 = 𝐿𝐿𝐹𝐹𝐹𝐹 + 𝐿𝐿𝐿𝐿𝐿𝐿 (50)

The international bond is in zero net supply internationally. Market clearing in the international
bond markets requires:
0 = ∑𝑟𝑟 𝐵𝐵 ∗ 𝑟𝑟 (51)

20
 Appendix B: Key Model Parameters

Definition USA CHN ROW

1/𝜎𝜎 Intertemporal elasticity of substitution 0.33 0.33 0.33

slc Share of liquidity-constrained households 0.25 0.35 0.30
θ Probability of survival 0.90 0.90 0.90
χ Productivity decline rate by age 0.95 0.95 0.95
v Habit persistence in consumption 0.40 0.40 0.40
σv Elasticity of substitution between capital and labor 0.85 0.85 0.85
σm Elasticity of substitution between imports and 2.00 2.00 2.00
domestic goods
σm Elasticity of substitution between imports and 1.50 1.50 1.50
domestic services
σw Elasticity of substitution between imports from 2.25 2.25 2.25
different countries (goods)
σw Elasticity of substitution between imports from 1.50 1.50 1.50
different countries (services)
σf Elasticity of substitution between goods varieties 7.70 7.70 7.70
σl Elasticity of substitution between labor varieties 7.30 7.30 7.30
δ Depreciation rate of capital 0.06 0.09 0.06
g-1 Long-term growth rate of the global economy 0.02 0.02 0.02
𝐼𝐼
𝜑𝜑 Capital adjustment cost parameter 5.00 5.00 5.00
𝜑𝜑𝑝𝑝 Price adjustment cost parameter 40.0 40.0 40.0
𝜑𝜑𝑤𝑤 Wage adjustment cost parameter 40.0 40.0 40.0
𝜔𝜔𝜋𝜋 Monetary policy weight on inflation gap 0.75 0.75 0.75
𝜔𝜔 𝑦𝑦 Monetary policy weight on output growth gap 0.50 0.50 0.50
Source: Kumhof and Laxton (2017), and author’s assumptions on the basis of related GIMF literature.

21
 Appendix C: Simulation Results of Trade War Scenarios (% change from the baseline)

Trump Tariffs China Retaliation Escalated Trade War

Year 1 Year 2 Year 3 Steady Year 1 Year 2 Year 3 Steady Year 1 Year 2 Year 3 Steady
State State State

Real GDP
USA -0.1 -0.8 -1.2 -2.9 -0.2 -0.9 -1.3 -3.0 -0.3 -1.2 -1.8 -4.1
China -0.5 -0.5 -0.4 -1.4 -0.6 -0.6 -0.6 -1.7 -0.8 -0.9 -1.1 -3.0
Rest of the world -0.5 -0.4 -0.4 -1.0 -0.4 -0.4 -0.3 -0.9 -0.4 -0.4 -0.3 -0.9
CPI inflation (percentage point changes)
USA 0.7 0.1 0.1 0.0 0.7 0.1 0.1 0.0 1.0 0.2 0.1 0.0
China 0.4 -0.1 0.0 0.0 0.4 0.0 0.0 0.0 0.7 0.0 0.1 0.0
Rest of the world 0.5 -0.2 0.0 0.0 0.4 -0.2 0.0 0.0 0.4 -0.2 0.0 0.0
Consumption
USA 0.8 0.8 0.7 -0.4 0.8 0.7 0.6 -0.7 0.8 0.6 0.4 -1.7
China -0.3 -0.4 -0.5 -2.1 -0.3 -0.4 -0.5 -2.3 -0.5 -0.7 -1.0 -3.9
Rest of the world -0.2 -0.2 -0.2 -1.8 -0.2 -0.2 -0.2 -1.7 -0.2 -0.1 -0.1 -1.7
Investment
USA -3.6 -5.4 -6.0 -3.9 -3.5 -5.4 -6.1 -4.1 -4.3 -6.7 -7.9 -5.5
China -0.6 -0.9 -1.2 -2.4 -0.6 -1.0 -1.3 -2.7 -1.0 -1.8 -2.3 -4.4
Rest of the world -1.5 -1.6 -1.4 -1.8 -1.4 -1.4 -1.3 -1.7 -1.4 -1.4 -1.2 -1.8
Exports
USA -17.1 -17.4 -17.0 -19.8 -17.6 -17.9 -17.6 -20.6 -20.5 -20.9 -20.5 -23.7
China -7.3 -3.9 -2.8 -2.9 -7.7 -4.9 -3.9 -4.5 -10.6 -7.2 -6.1 -7.5
Rest of the world -4.2 -1.9 -1.3 -1.1 -3.9 -1.8 -1.3 -1.1 -3.9 -1.7 -1.2 -1.0
Imports
USA -13.8 -12.2 -10.9 -5.0 -14.2 -12.8 -11.7 -6.5 -16.5 -15.2 -14.2 -8.5
China -9.2 -6.1 -5.4 -6.2 -9.5 -6.8 -6.2 -6.8 -13.4 -10.7 -10.3 -10.9
Rest of the world -4.3 -2.1 -1.6 -2.7 -4.0 -2.0 -1.5 -2.6 -4.0 -1.9 -1.3 -2.6

22
 Trump Tariffs China Retaliation Escalated Trade War

Year 1 Year 2 Year 3 Steady Year 1 Year 2 Year 3 Steady Year 1 Year 2 Year 3 Steady
State State State

Exports to the U.S.

China -47.2 -46.5 -45.8 -43.4 -47.7 -47.2 -46.8 -45.3 -79.7 -80.2 -80.5 -81.1
Rest of the world -11.7 -10.1 -8.8 -3.5 -12.1 -10.7 -9.6 -4.9 -8.9 -7.3 -6.2 -1.5
Imports from the U.S.
China -22.8 -22.8 -22.4 -25.0 -55.3 -55.3 -55.1 -56.5 -84.1 -84.1 -84.0 -84.5
Rest of the world -19.1 -19.7 -19.3 -21.8 -17.6 -18.2 -17.8 -20.3 -19.4 -20.1 -19.6 -22.0
Real exchange rate against the U.S. dollar
China 11.8 11.6 11.4 12.7 10.4 10.1 10.0 11.0 12.6 12.3 12.2 13.0
Rest of the world 9.1 9.1 8.9 10.5 8.3 8.2 8.1 9.6 8.9 8.8 8.7 10.3
Source: Model simulations.

23