China

2 sentences about Ethiopia · since 1946 · China profile
+30

Ethiopia

27 sentences about China · since 1946 · Ethiopia profile
Borders are for orientation only and say nothing about disputed territory.

Dashed outlines: areas whose status is disputed. Shapes follow Natural Earth, which draws some areas along recognised borders and others by who administers them on the ground; they say nothing about who is right, and front lines of current wars are not shown.

In short

  • Armed conflict: UCDP records them on opposite sides of the war in North Korea, South Korea, each supporting a side, most recently in 1953.Opposite sides
  • At the UN their vote agreement is 79% (2022–2024), about average for both of them.79%vote agreement
  • Trade in goods ($3.4 billion a year) is very important for Ethiopia, much less so for China.$3.4 billiona year
  • In its UN General Debate speeches, Ethiopia names China far more often than the reverse (27 sentences against 2).2 · 27sentences
  • 27,000 people born in China live in Ethiopia (UN DESA counts place of birth, however long ago they moved).27,000people
Written by fixed rules from the data below — no AI.

How the index is built

Each score runs from −100 to +100 and counts by its weight. The parts add up to the index.

UN votes 35% +7 score +19 · Vote agreement 79%High confidence
Trade 20% +19 score 97 · $3.4 billion a yearHigh confidence
Security 30% 0 score 0 · Opposite sides of armed conflict, last in 1953: too long ago to count under the rulesHigh confidence
Shared clubs 15% +4 score 25 · 1 shared groupHigh confidence
Words–Not scored until people check the stance labels (0 of 29 checked)
Index+30
What matters to you? Adjust the weights
Explain this scorehow each number is built, step by step

Every score follows fixed rules written out on How it works; no AI is involved. Here are the steps for this pair, with its own numbers. How it works · JSON · Downloads

UN votes +19

  1. Votes are compared with each country's own voting partners, not with all pairs: most of the world votes alike, so a global ranking would make France–Germany +62 and Canada–United States +6.
  2. Among China's 191 voting partners, Ethiopia ranks 29 (1 is the closest). That is position 85 on a scale from 0 (the furthest partner) to 100 (the closest).
  3. Among Ethiopia's 191 voting partners, China ranks 126: position 34.
  4. The two positions are averaged (60) and stretched to the range −100 to +100: +19.

Inputs: Vote agreement 79% on recorded votes, 2022–2024 (the same vote counts 1, an abstention against a yes or a no ½).

Trade 97

  1. Trade is scored by how much the more dependent side relies on it: the larger of the two shares of total trade, here 20.7%.
  2. On a log scale, 0.1% or less scores 0 and 25% or more scores 100 (a percentile would make 0.1% of a country's trade look like a middling tie).
  3. (log10(20.7%) − log10(0.1%)) ÷ (log10(25%) − log10(0.1%)) × 100 = 97, kept between 0 and 100, so the score is 97.

Inputs: Goods trade both ways: $3.4 billion a year, 2023–2025 (IMF) · 0.0684% of China's trade, 20.7% of Ethiopia's.

Security 0

  1. The Uppsala Conflict Data Program records the two on opposite sides of armed conflict. Each kind has a weight (main parties 100, troops 60, proxy 30) that fades with time: ×1 within 4 years of the last data year, ×0.5 for 5 to 14 years, ×0.2 for 15 to 29, 0 after that.
  2. Both only supporting a third party's side, last in 1953: 30 × 0 = 0
  3. The largest counts: +0. Recent fighting on opposite sides outweighs any alliance.

Inputs: Opposed: proxy 1953. Data to 2025. · Not both NATO members.

Shared clubs 25

  1. Each tracked group both belong to adds 25, and so does the same UN regional group. Capped at 100.
  2. 1 shared group (BRICS) = 1 × 25 = 25.

Inputs: Tracked groups: EU, AU, ARAB, ASEAN, BRICS, G7, G20, GCC, P5.

Index +30

ScoreValueWeightCounts forAdds
UN votes193535%+6.7
Trade972020%+19.4
Security03030%+0.0
Shared clubs251515%+3.8

The index is the weighted average of the scores that exist: (35×19 + 20×97 + 30×0 + 15×25) ÷ 100 = +29.8, shown as +30. Labels: 55 and up Very close, 25 Close, 0 Neutral, −30 Distant, below Opposed. Weights are yours to change above.

Words (what the two say about each other) are shown but not scored, and the people layer (migration, refugees, football) never is.

How sure are we?how much data stands behind each number

A score is only as good as the data under it. Four levels: High, Medium, Low, None. The thresholds are written on How it works; the level does not change a score, it tells you how far to trust it. How it works

UN votesHigh259 shared votes on whole texts in 2022–2024, the years the headline share covers. With that many, a share is good to about ±6 points.
TradeHighGoods trade is reported in 3 of the 3 years 2023–2025. Goods only, from customs data (IMF); services and energy pipelines are not in it.
SecurityHighThe Uppsala Conflict Data Program covers every state up to 2025. It counts only conflicts with at least 25 battle deaths a year.
Shared clubsHighThe list of tracked groups is fixed; membership is checked by hand.
IndexHighAs sure as the weaker of its two required parts, votes and trade. Security and shared clubs add to it but never make it surer.

Do their voting positions really differ?

In 2025, China stood at −0.37 (90% range −0.56 to −0.21) and Ethiopia at −0.29 (−0.47 to −0.11) on the main dimension of UN voting. The gap is 0.08. Their 90% ranges overlap: from their votes alone we cannot tell the two positions apart.

Positions are the ideal points of Bailey, Strezhnev and Voeten, estimated from every recorded vote; higher means voting more like the United States and Western democracies. They are not a moral score.

Words (what the two say about each other) are counted, not scored, so they carry no level. Moments and the people layer are context and never enter a score.

01 · The story so far

Eighty years, one line

How closely they voted at the UN (vote agreement). Bands are eras found in the data itself; dots are moments; below, who led each country.

Vote agreement on recorded General Assembly votes on whole texts: the same vote counts 1, an abstention against a yes or a no ½, a yes against a no 0. Source: Nations on Record (agreement score of Voeten et al.; UN Digital Library). Leaders: Wikidata, checked by hand.

1971–2016
90%
vote agreement
2017–2024
78%
Further apart than before

From the data

1951–1953UCDP records them on opposite sides of armed conflict in these years.
1971The first time Ethiopia named China in a UN General Debate speech.
1998The first time China named Ethiopia in a UN General Debate speech.
2015Their highest recorded trade in goods, adjusted for inflation: $5.1 billion both ways in 2025 US dollars ($3.8 billion in the dollars of the time).
Found in the data by four fixed rules, the same for every pair, not chosen by an editor. Never part of a score. The rules
  • Voting: the year with the biggest difference in vote agreement between the ten years before it and the ten years from it, if that difference is at least 15 points.
  • Armed conflict: every period of consecutive years in which UCDP records the two on opposite sides.
  • Trade: the year with the highest trade in goods both ways, adjusted for inflation with US consumer prices (BLS), if trade is recorded in at least five years.
  • Speeches: for each of the two, the first year it named the other in its UN General Debate speech, counted from the first year both voted at the UN (before that, a name can mean a territory or a state not yet independent).

02 · Say

In their own words

The conversation they never quite had: for each decade, what each government said about the other in its UN General Debate speech.

1960s
The League signed its own death warrant when it capitulated in face of the invasion of the Chinese north-eastern provinces—Manchuria—by the Japanese militarists, of Mussolini's rape of Ethiopia, and of Hitler's expansionist policies in Europe. Mr. Shen · China · 1961
It is obvious today that the United Nations cannot be effective in practical terms in dealing with the problems of war and peace in South-East Asia, for example, and disarmament, without the presence of the People's Republic of China in the United Nations. Mr. YIFRU · Ethiopia · 1967
1970s
My Government has expressed to the Government and people of China at the highest level the sense of loss and grief the Ethiopian people feels. Mr. Wodajo · Ethiopia · 1976
1990s
The border conflicts between Ethiopia and Eritrea and the conflict in the Democratic Republic of the Congo are another source of concern for us. Tang Jiaxuan, Minister for Foreign Affairs · China · 1998
2020s
China and South Africa has championed that call, and that is why Ethiopia is grateful to have been invited to join the group. Demeke Mekonnen Hassen, Deputy Prime Minister · Ethiopia · 2023

Quote rule: for each decade, the first sentence in the latest speech of that decade in which one names the other. Same rule for every pair and both sides.

Who talks about whom

ChinaEthiopia

Words next to the name

When China speaks

invasion1

When Ethiopia speaks

nuclear2
gratitude1
dialogue1

A keyword in the same sentence, not a judgement of who did what.

03 · Do

Votes, trade, war and alliance

Topic by topic

Most alike on Middle East (97% of votes the same), furthest apart on nuclear weapons (73%), against 88% on all votes since 2010.

Colonialism: the vote agreement fell from 92% (2010–14) to 81% (2015–19).

  • All roll-call votes 88% (range 87–90%) since 2010, 1406 votes · 90% all years 2010–14: 84% 2015–19: 78% 2020–25: 80%
  • Disarmament 80% (range 74–85%) since 2010, 178 votes · 83% all years 2010–14: 70% 2015–19: 62%
  • Middle East 97% (range 93–98%) since 2010, 188 votes · 97% all years 2010–14: 93% 2015–19: 93% 2020–25: 92%
  • Human rights 88% (range 83–91%) since 2010, 244 votes · 89% all years 2010–14: 80% 2015–19: 73% 2020–25: 74%
  • Colonialism 92% (range 85–96%) since 2010, 97 votes · 96% all years 2010–14: 92% 2015–19: 81%
  • Nuclear weapons 73% (range 66–78%) since 2010, 199 votes · 77% all years 2010–14: 62% 2015–19: 49% 2020–25: 58%
  • Economic development 93% (range 85–96%) since 2010, 94 votes · 94% all years 2010–14: 92% 2015–19: 87%

Vote agreement on roll-call votes, counted vote by vote the same way as the headline figure: the same vote counts 1, an abstention against a yes or a no ½, a yes against a no 0. Votes on amendments and single paragraphs are left out. Topics are the labels of the UN voting dataset; one resolution can carry several, and about 60% carry none. Disarmament, colonialism and economic development are coded together from 2019, so they count votes up to 2018. Bar: 2010–2025; mark: all years. Shares need at least five shared votes. The thin line on each bar is the 95% sampling range given the number of votes.

Votes that set them apart

Since 2010 they voted on 1406 whole-text resolutions together. On 49 one said yes and the other no; they differed in some way on 278. The 10 newest, newest first: (the API lists up to 50: API)

Roll-call votes on whole texts at the UN General Assembly, since 2010; amendments and single paragraphs are left out, as in the topic bars. Titles and document symbols: UN Digital Library via Nations on Record; votes: Voeten et al. A vote is not a statement about the other country, only about the text.

Voting positions over time

Each year's position on the main dimension of UN voting, with its 90% range. Where the shaded ranges overlap, the votes of that year cannot tell the two apart. The gap was 0.72 in 1971 and is 0.08 in 2025; it was widest in 1995 (1.52).

-2.5-2.0-1.5-1.0-0.50+0.5 19801990200020102020

China Ethiopia

Show data table
YearChina90% rangeEthiopia90% range
2025-0.37-0.56 – -0.21-0.29-0.47 – -0.11
2024-0.86-1.02 – -0.7-0.68-0.88 – -0.48
2023-0.85-1.02 – -0.68-0.54-0.72 – -0.35
2022-1.07-1.26 – -0.88-0.97-1.16 – -0.77
2021-0.41-0.58 – -0.24-0.7-0.95 – -0.47
2020-0.59-0.79 – -0.38-0.47-0.72 – -0.25
2019-0.45-0.63 – -0.27-0.44-0.66 – -0.23
2018-0.51-0.69 – -0.34-0.35-0.54 – -0.17
2017-0.31-0.49 – -0.14-0.54-0.81 – -0.29
2016-0.49-0.69 – -0.29-0.53-0.8 – -0.26
2015-0.26-0.45 – -0.11-0.54-0.79 – -0.3
2014-0.81-1.06 – -0.56-0.56-0.83 – -0.33
2013-0.6-0.82 – -0.37-0.54-0.82 – -0.29
2012-0.9-1.15 – -0.65-0.42-0.68 – -0.2
2011-0.57-0.81 – -0.34-0.38-0.61 – -0.17
2010-0.85-1.14 – -0.57-0.45-0.71 – -0.22
2009-0.57-0.8 – -0.35-0.42-0.65 – -0.2
2008-0.6-0.86 – -0.36-0.45-0.67 – -0.23
2007-0.5-0.73 – -0.29-0.47-0.74 – -0.24
2006-0.73-1.01 – -0.47-0.34-0.54 – -0.16
2005-0.68-0.94 – -0.44-0.36-0.56 – -0.18
2004-0.75-1.1 – -0.44-0.45-0.72 – -0.21
2003-0.92-1.24 – -0.64-0.23-0.44 – -0.04
2002-0.64-0.93 – -0.36-0.3-0.5 – -0.11
2001-0.55-0.81 – -0.31-0.55-0.84 – -0.29
2000-0.73-1.02 – -0.46-0.37-0.63 – -0.13
1999-0.6-0.85 – -0.34-0.41-0.66 – -0.18
1998-0.39-0.62 – -0.16-0.33-0.59 – -0.08
1997-1.21-1.54 – -0.89-0.11-0.32 – 0.09
1996-1.22-1.57 – -0.86-0.27-0.53 – -0.03
1995-1.65-2.01 – -1.3-0.13-0.38 – 0.11
1994-1.52-1.9 – -1.14-0.44-0.76 – -0.13
1993-1.42-1.78 – -1.03-0.21-0.51 – 0.03
1992-0.63-0.95 – -0.32-0.32-0.58 – -0.07
1991-0.39-0.7 – -0.1-0.41-0.7 – -0.16
1990-0.12-0.37 – 0.09-0.63-0.98 – -0.3
1989-0.04-0.26 – 0.17-0.71-1.08 – -0.36
19880.01-0.17 – 0.19-0.93-1.25 – -0.61
19870.170.01 – 0.32-0.86-1.14 – -0.59
19860.180.05 – 0.31-0.9-1.12 – -0.66
19850.20.06 – 0.32-0.71-0.89 – -0.51
19840.190.07 – 0.32-0.75-0.97 – -0.55
19830.180.05 – 0.32-0.78-1.04 – -0.54
19820.14-0.01 – 0.27-0.99-1.25 – -0.75
1981-0.08-0.27 – 0.11-0.81-1.05 – -0.58
1980-0.02-0.25 – 0.18-0.78-1.0 – -0.56
1979-0.11-0.36 – 0.13-0.5-0.73 – -0.27
1978-0.12-0.51 – 0.18-0.49-0.74 – -0.26
1977-0.84-1.38 – -0.33-0.3-0.67 – 0.04
1976-0.99-1.34 – -0.64-0.4-0.78 – -0.04
1975-0.2-0.56 – 0.11-0.03-0.32 – 0.24
1974-0.88-1.49 – -0.33-0.18-0.57 – 0.17
1973-0.09-0.55 – 0.34-0.24-0.71 – 0.19
1972-0.48-1.01 – 0.040.0-0.39 – 0.36
1971-0.74-1.43 – -0.13-0.02-0.39 – 0.35

Positions are the ideal points of Bailey, Strezhnev and Voeten, estimated from every recorded vote; higher means voting more like the United States and Western democracies. They are not a moral score. Source: Nations on Record (Bailey, Strezhnev and Voeten).

On the UN voting map

In 2020–24, China voted with Russia, Iran and 7 others (10 countries in all) and Ethiopia with India, Indonesia and 60 others (63). They shared a bloc in 1 of the 6 periods since 1946.

Loading the map…

Drag the year to see how the blocs changed since 1946.

Each dot is a country, placed by how alike it voted that year: close dots voted alike. Colours are voting blocs, groups of countries among each other's closest voting partners in that period. The axes have no units; the United States is kept on the right and China at the top. Click a dot to open that country's pair with China. Source: Nations on Record, which shows the blocs in full. Inside the blocs (Nations on Record). Never part of a score.

Friends in common

The countries each one votes most alike with at the UN. Countries in the middle vote alike with both.

Trade over time

China sells to EthiopiaEthiopia sells to China

Goods exports, current US dollars. IMF via Nations on Record.

Who needs whom

Arrow width: yearly goods flow in each direction (2023–2025). Bars: share of each country's total trade.

Security

  • 1951–1953North Korea, South Korea — each supporting opposite sides of a third party's war

Counted: the two governments on opposite sides, or one sending troops to fight against the other. Fighting with armed groups based in the other country is not counted. UCDP Dyadic Dataset 26.1, data to 2025.

Shared clubs

BRICS

EU, African Union, Arab League, ASEAN, BRICS, G7, G20, GCC, the P5 and UN regional groups. NATO counts under Security.

Trade agreements

No trade agreement between them is notified to the WTO.

Agreements notified to the WTO, in force or signed. When a bloc such as the EU signs as one party, the agreement links its members with the other side, not with each other. The status is as notified to the WTO; whether an agreement is still applied between these two countries is not checked. Source: WTO Regional Trade Agreements database.

Sanctions

No UN sanctions apply to either. Sanctions the two countries may impose on each other themselves are not tracked here.

Country regimes of the UN Security Council, the EU, the United States and the United Kingdom; thematic lists (terrorism, cyber, human rights) are left out, and so are other countries' own lists. A regime can target a government, officials or companies; it is not a measure of the relationship. Sources: EU Sanctions Map, OFAC (US Treasury), UK government. Never part of a score.

04 · Live

Between people

Governments argue; people move, study, flee and play football. None of this enters the index.

Living in the other country

born in Chinaborn in Ethiopia

International migrant stock, UN DESA 2024. By place of birth: this includes refugees, and people born before today's borders, for example in a state that later split.

Refugees & asylum seekers
0
from China in Ethiopia
19
from Ethiopia in China (2025)

UNHCR. Refugees and asylum seekers only: people given citizenship on arrival are not counted.

Ask about China & EthiopiaLocal AI · answers only from this page's data · numbers checked

The answers are written by an AI language model, not by the site's editors. It can be wrong; the numbers and sources on this page are the reference.

What changed recently

  • Voting together: about the same (79% in 2022–2024).
  • Trade: up 44% ($2.3 billion → $3.4 billion a year).
  • Sentences naming each other at the UN: 1 in the last five General Debates, 0 in the five before.

Sources & coverage

  • Speeches, mentions, votes, trade: Nations on Record, loaded 2026-10-12 20:33 UTC.
  • UCDP Dyadic and Battle-Related Deaths datasets, version 26.1, loaded 2026-11-14 20:06 UTC.
  • UN DESA International Migrant Stock 2024, loaded 2026-10-12 20:18 UTC. UNHCR Refugee Population Statistics (api.unhcr.org), 2000–2025, loaded 2026-10-12 20:18 UTC.
  • International football results 1872–present (Mart Jürisoo, CC0), loaded 2026-10-12 20:18 UTC.
  • Scoring rules 0.4 (14 Nov 2026), scores built 2026-11-14 21:20 UTC. Every rule.