Malaysia has known this for fifty years. There is no export-led economy until salaries grow.
Ethiopia's merchandise exports went from 3.8 billion US dollars in 2023/24 fiscal year to 8.3 billion US dollars in 2024/25 fiscal year to roughly 10.7–11 billion US dollars in 2025/26 fiscal year. The government has presented this as evidence of structural transformation. The number is real. It is also not a description of what an Ethiopian household earns, spends, or keeps, and the gap between those two facts is worth sitting with before the next set of figures gets announced.
Look at where the growth actually came from. Coffee and gold produced 78 percent of export revenue this year, a share that is rising rather than falling, moving in the opposite direction from the government's own diversification goal even as the headline number improved. Coffee's price moved from roughly 2.83 US dollars a pound in 2023 to 3.12 US dollars in 2025. Gold peaked near 2,787 US dollars an ounce in October 2024, a 27 percent jump that landed in the same window as Ethiopia's own rise in output. Both effects are real. Neither is a story about productivity, and neither is guaranteed to repeat.
Underneath the export line sits a currency that has moved further than the exports themselves. The birr has depreciated roughly 181 percent against the dollar since the July 2024 float. Every dollar of coffee or gold revenue converts into more birr than it did a year earlier, independent of anything grown or mined. Reserves stood at 4 billion US dollars to 4.4 billion US dollars by mid-2026, covering 1.8 to 2.5 months of imports against a conventional three-month threshold. Inflation, which had fallen to single digits by January 2026 for the first time in years, rose back to 11.4 percent by April. The clearest available wage benchmark, the Birches Group's regional survey, found private-sector wage adjustments of roughly 20 percent against inflation running near 26 percent over the same period — a real-terms loss for salaried Ethiopians during the highest-inflation months of the transition.
Malaysia already ran this experiment, and published the results. Its export-led model dates to the 1970s and 1980s: an aggressive push into foreign-direct-investment manufacturing, mostly electronics assembly, layered on an older base of tin, rubber, and petroleum through the state oil company Petronas. The World Bank's 1993 account of the "East Asian Miracle" held Malaysia up as the model of the type. Growth followed for two decades. The distributional record, read forward from 1970, tells a narrower story than the growth headline.
Malaysia's Gini coefficient fell from 0.513 in 1970 to 0.411 in 2020. That is genuine progress, and it is also a plateau: the coefficient reached an official low near 0.40 around 2014 and has barely moved since, across a further decade of continued growth. Compensation of employees — the share of GDP paid out as wages rather than kept as corporate profit — stood at 33.6 percent in 2024, against a government target of 40 percent unmet for over a decade, and against shares above 50 percent in the United States and Germany. The World Bank's 2026 Malaysia Economic Monitor found real median wages rose 43 percent between 2010 and 2024, roughly half the pace of GDP growth over the same period. Former Bank Negara governor Muhammad Ibrahim put a sharper number on it in 2025: comparing his own 1984 starting salary to a 2025 graduate's, adjusted for four decades of inflation, he estimated real purchasing power for entry-level wages had fallen by a factor of roughly three.
The two records point to the same transmission problem rather than two different ones. In both economies, foreign earnings accrue first to a narrow set of institutions positioned at the export interface — Petronas and Malaysia's government-linked companies in one case, licensed coffee and gold exporters and FX-linked trading houses in the other — before any of it reaches a wage packet. What happens next runs through a separate and slower set of channels: minimum wage law, collective bargaining coverage, fiscal transfer design, and how much of a weaker currency passes through into the price of what a household actually buys. These channels do not move on the export data's clock. Malaysia's compensation share sat below 34 percent of GDP in 2024, fifty years after its own export boom began. That is the empirical answer to how long a distributional gap can persist without correcting itself through growth alone.
The parallel has a limit worth stating rather than smoothing over. Malaysia's inequality carries an explicit ethnic-political architecture that Ethiopia's data does not show — the New Economic Policy's race-based redistribution targets and four decades of contested affirmative-action politics have no equivalent in the Ethiopian record. Malaysia's labour market also carries a second feature Ethiopia has not yet built at comparable scale: a foreign workforce of 2.37 million people, 14 percent of the total, concentrated in low-wage sectors that suppress wage growth at the bottom by design, according to Malaysia's own former central bank governor. Ethiopia's transmission problem, on the evidence available, runs through currency-float design and commodity concentration rather than through an explicit quota system.
There is also an asymmetry in how long each country has been measuring itself. Ethiopia is two years into a currency float. Malaysia has fifty years of documented data, produced by a Household Income and Expenditure Survey run continuously since 1970. That survey is the exact instrument Malaysia used to discover, decades into its own boom, that the aggregate and the median household had stopped moving together. Ethiopia has not yet built an equivalent at the same frequency: no current wage index, no updated employment survey, no household consumption data published at the pace the export figures arrive. That is not a minor omission. It is the precondition for asking Malaysia's question at all.
An economy is a set of accounts. A people is not an account, and the distance between the two is measured only by data a government chooses to collect, publish, and defend against its own good news. Malaysia collected that data for fifty years, and the resulting picture showed compensation still below a third of GDP and a Gini coefficient that had improved by less than a fifth of a point since 1970, despite five decades of growth described at every stage as a national success. Ethiopia's reporting is, so far, honest about the export and currency figures and candid about what is missing. That candor is worth more than a single quarter's growth number. The figure that matters is not this year's export total. It is whether, ten years from now, Ethiopia has built the statistical apparatus to answer the question Malaysia only got around to asking after fifty.
Bayu Samuel is a Strategy & Investment Analyst with experience across Ethiopia’s public and private sectors. He holds a BSc from Addis Ababa University and applies expertise in securities, financial modeling, and game theory to guide investment decisions. He previously served as a Senior Analyst at the FDRE Land Bank and Development Corporation.
