Initially, mine workers would be rather reluctant to invest their wages in means of production (in agriculture and in transport) within the Mozambican rural economy. Up to 1980/81, government policies were not favourable to such investments. However, thereafter, miners were specifically encouraged to plough back their wages into production and commerce. Rural unemployment was widespread and, hence, the conditions for private accumulation were favourable on this count. Generally, miners would invest in transport and commerce, but some did invest in agriculture. Indeed, in the latter years, peasants with resources were allowed to operate on unutilised ex-settler farms. In other cases, the more permanent and better paid state farm workers could use their specific position to strengthen their own farm, often supplemented by hired labour. As mechanics or tractor drivers, etc. they had access to cer-tain resources such as seeds, fertiliser, fuel and consumer goods which they could buy either from the state farm or, not unfrequently, merely take from stocks on the state farms. Border areas were another such case of differentiated access to resources by means of barter trade cross the border. Due to the political criticality of such areas within a general condition of war, the government distribution policy would grant a certain priority to supplying these areas with commodities which would then provide a basis for further barter trade with the neighbouring country. Further, areas located more closely to the main food markets (either towns or plantations) would be subject to a much more dispersed and intensive barter and money trade, thereby raising the producer prices which would benefit those peasants who had sufficient resources to produce surpluses. More distant food producing areas were much more within the grip of the commercial traders who provided the link with the market. Hence, while some strata within the peasantry managed to create some room for themselves by producing for the parallel markets, the majority of rural producers (either as wage labourers or small-scale producers) confronted declining real incomes as a result of the inflation on the parallel markets to which they had to turn not only for industrial commodities but also to supplement their food needs. Hence, their problem was not one of having too much money at hand with too few commodities to buy; rather, they experi-enced an acute shortage of both money and goods. The poorer peasantry were the main suppliers of seasonal labour to the state sector. However, although rural unemployment was high, the supply of labour was by no means elastic. The reasons for this were the following. First, the pattern of labour demand of the state farms and plantations was in most cases highly seasonal and, hence, did not provide an all-round income for the worker. Second, money wages earned on the state farm did not guarantee any access to commodities, and often did so only at speculative prices. For both reasons, the real basis of security of the rural worker still remained his family farm, however fragile that may have been. The state sector may have become dominant in terms of area and in terms of production (regarding monetary output), but it certainly was not the dominant aspect in securing the livelihood of rural producers. In most cases, the pattern of peak demand for labour on the state farms coincided with the peak demand for labour in family agriculture. For example,