The Asian financial crisis led to a major devaluation of the Indonesian exchange rate, macro instability, and the need for a “structural adjustment” program. The real devaluation affects prices throughout the economy and has a major impact on growth, production, deforestation, and income distribution in the Sumatera region. This paper uses computable general equilibrium (CGE) models —a national model and a regional model of Sumatera— that focus on agriculture to explore the impact of a real devaluation on the economy of Sumatera. The model incorporates commodity and factor market linkages between Sumatera, the rest of Indonesia, and the world (through commodity trade). Simulations are conducted for the short and medium run under alternative scenarios of macro adjustment. Structural adjustment causes Sumatera's economy to contract in both the short and medium run. Devaluation leads to increases in the prices of tradable goods. Regional exports increase, mainly from the non-agricultural sectors, and imports decline. The agricultural terms of trade (agricultural prices relative to non-agricultural prices) decline because agriculture has a small share in both regional and national exports. Food crops, except for rice and sugar, decline while tree crops increase. Deforestation is likely to increase because demand for forestry products increases, both as final products and as intermediate goods for the wood processing industry, both of which are sold on international markets. We analyze a possible policy response of imposing an export tax of 5-20 percent on processed wood to discourage further deforestation in the region. The results show that the proposed export tax reduces production of raw timber and processed wood, but at the cost of lowering exports and hence making the macro adjustment more difficult. Given the current situation, it is impossible to predict exactly how the resolution of the current macro crisis will unfold. We model two alternative macro adjustment scenarios that should bracket the likely response of the Sumatera region to the devaluation and structural adjustment program. In the first, we assume that regional investment is a proportion of regional aggregate income (or regional absorption), and that the adjustment burden is shared proportionately between aggregate consumption and investment. In the second, regional savings and investment are assumed to fall more, as the region’s trade balance is forced to improve dramatically to reflect the large required changes in the national trade balance. Under the first adjustment scenario, a devaluation benefits large farm households in the rural area and urban high income households who are involved in the production of tradable goods, while both rural poor and urban poor households lose. When investment falls more, all farm households and urban high income households benefit, but poor urban households lose more because the decline in investment hurts unskilled urban labor, especially in construction.