How does globalization affect domestic inflation?

The short answer

For four decades, central bankers have wrestled with one of the great macro mysteries: why did inflation stay low and stable across most advanced economies despite tight labor markets and aggressive monetary easing? The Phillips curve — the historical relationship between unemployment and inflation — flattened dramatically after 1990, leading some researchers to declare it dead.

One leading explanation is globalization. As China, Eastern Europe and other emerging economies integrated into global trade, the effective global labor supply roughly doubled. Cheap imports kept goods prices low. Global value chains meant that even non-traded services were exposed to international cost competition through their inputs.

The angle that distinguishes the modern literature: it’s not trade openness in the simple sense (imports + exports / GDP) that matters most. It’s the depth of integration via GVCs — how many production stages cross borders. This subtler measure, developed by Auer, Borio and Filardo at the BIS, explains the Phillips curve flattening better than headline trade ratios.

→ New to international economics? Financial education hub

What the data shows

The most cited evidence comes from the BIS Working Paper 602 (Auer, Borio and Filardo, 2017) and subsequent updates by Mikolajun-Lodge and others.

The empirical context (BIS, IMF, BLS, 1990-2024):

• Global trade as % of GDP: from ~39% in 1990 to ~60% pre-COVID, falling to ~57% in 2023
• Goods CPI in advanced economies: averaged ~1% annually 1995-2019; services CPI averaged ~2.5%
• Goods-services inflation gap: roughly 1.5 percentage points sustained over 25 years
• BIS Phillips curve estimates: slope coefficient on output gap declined by roughly half between 1990s and 2010s in OECD economies
• China share of world manufacturing exports: from ~3% in 1995 to ~20% in 2020
• Post-COVID reversal: US goods inflation rose to 13% YoY at peak (June 2022) before normalizing

The exception worth noting: services inflation, which is less exposed to direct international competition, has remained more sensitive to domestic labor markets. The 2022-2024 inflation episode was concentrated initially in goods (supply chains, energy) before propagating into services through wage pressures.

→ Dataset: US Core CPI Inflation

Why it happens — the macro mechanism

Globalization affects domestic inflation through three reinforcing channels.

Channel 1 — Direct import competition. When domestic producers face cheap imports, their pricing power erodes. They cannot raise prices much above international competitors without losing market share. This is the most visible channel — and it shows up clearly in goods CPI being persistently lower than services CPI in nearly every advanced economy since 1995.

Channel 2 — Global value chains. Even producers of non-traded goods import inputs from abroad. A French car maker uses Korean batteries, Chinese electronics, German steel, Mexican wiring. When global input prices fall, domestic production costs fall — softening the link between domestic labor markets and final prices. The angle that distinguishes the BIS view: GVCs explain the Phillips curve flattening better than aggregate trade openness because they capture indirect competition. See our FAQ on imported inflation.

A third channel runs through wages.

Channel 3 — Wage competition and migration. When firms can credibly threaten to offshore production, domestic wage demands moderate. When low-wage immigrants enter local labor markets, the effective domestic labor supply expands. Both effects flatten the wage-price spiral. Bidner-Eswaran-Kotwal estimate that the global labor force entry of China and India added roughly 1.5 billion workers to the world economy between 1990 and 2010 — effectively doubling globally available labor. See our FAQ on wage-price spirals.

Synthesis by regime. In the hyperglobalization era 1990-2010, China’s WTO accession (2001), Eastern Europe’s EU integration (2004) and the broader expansion of GVCs combined to deliver persistent goods disinflation across advanced economies — Phillips curves flattened and central banks worried about deflation. From 2010 to 2019, GVC growth slowed but disinflation persisted, supported by zero-bound monetary policy and weak demand. Since 2020, supply chain disruptions, US-China decoupling, and reshoring have reversed parts of the structural disinflation — the 2022 inflation surge was not just monetary but also reflected fragmenting supply chains. The pivot is whether deglobalization makes future inflation structurally higher.

The Phillips curve did not die — it was outsourced. The question now is whether globalization is being recalled.

→ Conceptual framework: Macro-financial regimes

What it means for different economic actors

Bond investors. If globalization-driven disinflation reverses, the structural underpinning for low inflation breakevens weakens. Post-2022 breakeven inflation rates settled around 2.3-2.5% in the US, higher than the pre-COVID 1.7-2.0% range, possibly reflecting partial reversal.

Equity investors. Sectors with globally integrated supply chains (consumer electronics, autos, apparel) have benefited most from input cost compression. A reversal would compress margins for these sectors and benefit domestically focused producers.

Central banks. A flatter Phillips curve made it harder to overheat the economy in the 1990s-2010s; a steeper Phillips curve post-2020 means even modest tightening can produce meaningful inflation results. The Fed’s 2022-2023 hiking cycle delivered faster disinflation than 1970s tightening for partly this reason.

A common error is to treat the post-2022 inflation episode as definitive proof that globalization disinflation is over. The structural goods-services inflation gap remained, and supply chain normalization has materially eased goods inflation since 2023.

Practical observation

Go deeper

Related questions

• What is total factor productivity and why has it slowed?
• What is secular decline in real interest rates?
• What is the savings glut thesis?
• Why does inflation come in waves?

Frequently asked questions

Why do GVCs explain the Phillips curve flattening better than trade openness?

Simple trade openness (exports + imports / GDP) measures how much of total economic activity crosses borders, but it doesn’t capture the depth of integration. Two economies with identical trade ratios can have very different exposure if one ships finished goods and the other participates in multi-country production. GVC measures (developed by Koopman-Wang-Wei and used by the BIS) capture how many stages of production cross borders — and this depth correlates more strongly with the loss of pricing power that flattens domestic Phillips curves. Auer, Borio and Filardo (2017) document this empirically across OECD economies.

Did globalization really lower US inflation persistently?

The evidence is consistent but not unanimous. Goods CPI in the US averaged ~1% annually from 1995 to 2019, while services CPI averaged ~2.5% — a sustained gap of 1.5 percentage points that broadly aligns with the globalization channel. Laurence Ball (2006) argued the effect was small for the US specifically, but Auer-Borio-Filardo and others find it material once GVC integration is properly measured. The disagreement reflects genuine difficulty in isolating one structural force from many.

Is post-COVID reshoring reversing globalization-driven disinflation?

Partially. Goods inflation surged in 2021-2022 as supply chains broke and shipping costs spiked. Since 2023, much of that has reversed — US goods CPI is back near zero in 2024-2025. But the structural environment has shifted: US-China tariffs, EU carbon adjustments, and active reshoring policies are creating frictions that did not exist in 2010. Whether this produces structurally higher inflation or merely higher inflation volatility remains an open question.

Last updated — 12 July 2026