The views expressed in this Working Paper are those of the author(s) and do not necessarily represent those of the IMF or IMF policy. Working Papers describe research in progress by the author(s) and are published to elicit comments and to further debate. WP/02!75In a sample of 60 developing countries, we find evidence of a strong-almost one-toone-relationship between capital inflows and domestic investment. However, this relationship has evolved over time. While growing financial integration with the rest of the world has increased access to foreign private capital, the relationship between foreign capital and domestic investment has weakened, reflecting changes in the composition of inflows, offsetting outflows, and increased foreign-currency reserve requirements. In contrast, better policies have not only brought in more capital but also, especially for foreign direct investment, have tended to strengthen the relationship between foreign capital and domestic investment.

We examine the capital flows-domestic investment relationship for 60 developing countries from 1979 to 1999. In the 1990s, even as liberalization attracted new flows, foreign capital stimulated less domestic investment than in the preceding decade. With greater financial integration, governments accumulated more international reserves and domestic residents diversified by investing abroad. Foreign investors were also motivated by diversification objectives rather than by unmet investment needs. Inflows were channeled increasingly through portfolio flows-or through foreign direct investment with the characteristics of portfolio capital-resulting in weak investment stimulus.However, stronger policy environments strengthened the link between inflows and investment.

No abstract

If the food sector is attacked, the likely agents will be chemical, biological, or radionuclear (CBRN). We compiled a database of international terrorist/criminal activity involving such agents. Based on these data, we calculate the likelihood of a catastrophic event using extreme value methods. At the present, the probability of an event leading to 5,000 casualties (fatalities and injuries) is between 0.1 and 0.3. However, pronounced, nonstationary patterns within our data suggest that the "reoccurrence period" for such attacks is decreasing every year. Similarly, disturbing trends are evident in a broader data set, which is nonspecific as to the methods or means of attack. While at the present the likelihood of CBRN events is quite low, given an attack, the probability that it involves CBRN agents increases with the number of casualties. This is consistent with evidence of "heavy tails" in the distribution of casualties arising from CBRN events.

No abstract