How much the ariary gained against the US dollar in the year to end-March 2026 (BFM). For a post-transition economy that lost duty-free US access and took two cyclones in February, a strengthening currency is the most counter-intuitive number on the board and this edition is partly the story of why, and what it costs.

Several of this year's better-looking numbers trace back to a single cause. Late in 2025, large flows of foreign currency tied to development projects entered the country. Per the central bank (BFM), those inflows did three things at once: they pushed the ariary up (+8.9% against the dollar and +10.7% against the euro to end-March), they left commercial banks holding an "exceptionally high" level of excess cash, and they gave the BFM room to hold its key rate at 12% on 5 May while inflation eased to 6.1% in January before edging back to 6.8% in March. Foreign reserves now cover 7.3 months of imports a comfortable cushion by regional standards.

The benign reading is recovery, and part of it is real: the reserve buffer and the easing in inflation are genuine. But development inflows are lumpy and largely one-off, they are not export earnings, and exports are under pressure, not building. A stronger ariary cuts both ways: it helps importers and anyone carrying dollar costs, but it makes vanilla and textiles dearer abroad exactly as US tariffs already price them out. The strength is real but resting on one pillar.

Worth watching: whether the ariary holds once the inflows taper, and whether the idle bank cash reaches businesses a cushion buys time, not a turnaround, and the question is whether the time gets used.

The National Assembly passed a bill this month to widen credit access for small businesses. The intent is sound, but the timing exposes the real constraint. Banks are not short of money they are sitting on excess liquidity, and credit was already growing 10.3% year-on-year in March, concentrated in medium- and long-term lending to borrowers banks already trust.

When a system this liquid still leaves small firms unfunded, the bottleneck is not the supply or price of money; it is collateral and risk. Whether the law moves the needle turns on one question its headline does not answer: does it change what counts as collateral?

Worth watching: bank lending terms in the months after it takes effect if terms hold steady, the law is a signal, not a shift.

The visible story is recovery. Ambatovy, the nickel-and-cobalt complex that is the country's largest mining operation restarted its first acid plant on 23 May, after Cyclone Gezani damaged its Toamasina processing facility in February and shut it for about three and a half months. The second plant is due back by end-June, with a June target near 2,500 tonnes of nickel and 250 tonnes of cobalt, against a 2024 base of roughly 28,000 tonnes of nickel a year. It lands as the government courts miners: in January it lifted a 16-year freeze on new mining permits, reopening nickel, cobalt, graphite, rare earths, mineral sands and iron ore.

The quieter signal points the other way. In May, Sumitomo the Japanese trading house that anchored Ambatovy divested its 54% stake, leaving the Korean state minerals entity KOMIR as majority owner and the one funding the rebuild. The contrarian read is ordinary portfolio rotation: trading houses cycle assets, and KOMIR doubling down suggests at least one strategic buyer values non-DRC cobalt enough to stay. But the timing is hard to ignore the country's largest foreign miner left the same season a cyclone exposed the climate risk of island processing, and the same year the welcome mat went out.

Worth watching: who takes the reopened permits. If the buyers are mostly state-backed strategic players rather than commercial miners, the risk-adjusted returns only work for those buying for reasons other than profit.

In December the Ministry of Trade liberalised vanilla pricing, citing saturated stocks at home and abroad. It kept guideline export floors roughly USD 15/kg for powder, USD 25 for red, USD 50 for gourmet black below which beans cannot legally be exported. This is the second retreat from hard price control since the USD 250 floor collapsed in 2023, and "reference" floors are still floors: they hold only if buyers cannot source elsewhere. With global stockpiles full and the 2026 green season now opening, the risk is that floors set on paper push volume away rather than lift farmer income.

Worth watching: export volumes through the third quarter a drop would mean buyers are waiting out the price, not paying it.

The number everyone cites is the 47% US tariff announced in April 2025. It overstates the average pain and obscures who carries it. The rate actually collected on Malagasy goods averaged around 16% in early 2026 (US customs data), because the burden is split by product. Textiles duty-free under AGOA until it expired on 30 September 2025 are hit hard, with industry bodies putting tens of thousands of formal jobs at risk. Vanilla and minerals, which historically entered at low or zero duty, are far more shielded, and miners are petitioning for a critical-mineral exemption. The strong ariary compounds the squeeze on the one exposed sector that employs urban formal workers.

Worth watching: the stopgap one-year AGOA extension being lobbied in Washington the difference between managed decline and collapse for the textile zones.