Discussions of international security issues have largely been the province of political scientists, historians and specialists in international relations. Can economic analysis provide insights into questions of international security? With funding from the John D and Catherine T MacArthur Foundation, CEPR organized a series of workshops between November 1987 and March 1988 in order to encourage economists to explore security issues and in order to develop links between the different disciplinary approaches to international security. Each of the four workshops focused on a specific theme: the international debt crisis; international policy conflict and coordination; East-West economic relations; and conflicts over resources and agriculture.
The first meeting of the workshop series, on 20 November 1987, opened with a paper by Giacomo Luciani (UCLA and Institute for International Affairs, Rome) on `The Economy and National Security'. Luciani surveyed the different links between economic issues and security. He contrasted narrow and broad definitions of national security. The narrow definition sees security as concerned with the defence of the territorial integrity of a country against military attack. Broader definitions allow both for the defence of a range of `essential values', which extends beyond territorial integrity, and for threats to these essential values other than military attack. Within the narrow definition, the importance of the economy to security is well defined and fairly obvious; but as the definition is broadened, almost any economic event can be seen as having security implications. The wider definition of security may be so broad as to have little real content. The importance of the economic factors to security (narrowly defined) is also problematic: as the relative cost of modern weapons has decreased, the importance of economic power to military success may have decreased. Vietnam and Afghanistan provide two examples of the success of the economically weaker protagonist in a conflict. Luciani considered some of the uses to which the broader concept of the link between economics and security has been put. One is to justify a preference for domestic over foreign sources of supply for a wide range of strategically important goods, but Luciani argued that it was not clear that dependence on high-cost domestic supplies enhanced a nation's security. Luciani concluded that, although there was some positive value to a broad view of the links between economics and security, in most cases the application of the term `security' to normal economic or commercial risks was merely a rhetorical device.

Discussion of Luciani's paper showed a broad consensus among workshop participants that there is a very important set of issues linking economics and international security, such as the economic implications of defence spending, the use of economic pressure for political purposes, and the security implications of successful international economic cooperation. There was also broad agreement, however, that it was likely to be impossible to provide a general conceptual framework which could encompass all of these issues.

Security Aspects of the International Debt Problem
The remaining three papers at the first workshop were concerned with aspects of the international debt problem. Kenneth Kletzer (Yale University), in `International Policy and Bargaining Between Creditors and Sovereign Debtors', analysed the best response of a creditor to the situation in which a sovereign debtor is in danger of defaulting on its debt. In particular, Kletzer investigated whether it is ever desirable for lenders to extend additional funds to a sovereign debtor in order to reduce the probability of default. He also discussed the `public good' aspects of debt negotiations: would individual creditors find it optimal to undertake actions which are in the interests of creditors as a group? Kletzer presented a formal model in which both lenders and borrowers have the same information about the borrowers' true economic circumstances. In this situation it may be optimal for lenders to take steps to avoid defaults by partial write-offs which reduce borrowers' debt-service obligations, but not by rescheduling debt or extending new loans. When lenders are imperfectly informed about borrowers' true circumstances, however, offers of debt renegotiation are constrained by the need to ensure that the offers are only attractive to debtors who are genuinely in need of relief, and it may then become optimal to offer debt rescheduling and new loans. This argument, based on asymmetry of information, offers a very different justification of rescheduling or `defensive' lending to that in the existing literature on the debt problem.
In the discussion, Willem Buiter (LSE and CEPR) pointed out that the seniority ranking of new and old creditors played a key role in Kletzer's model, but that it did not seem to be a prominent feature of real-world discussions of rescheduling. He also observed that the question of the observability of different countries' `true' probabilities of default is actually a question about the preferences of governments.
Shanti Chakravarty (University of Wales, Bangor) asked `Why Is There No Debtors' Cartel in Latin America?' He observed that a large proportion of sovereign debt is owed by a small group of countries and that the debt is also concentrated in the portfolios of a small number of banks. The key question in the long run is whether countries can earn a rate of return on the resources they borrow that is higher than the interest rate applicable to the borrowing. Using a stylized growth model that incorporated foreign borrowing, Chakravarty argued that neither changes in the interest rate on borrowing nor changes in the level of outstanding debt have significant effects on growth in debtor countries. He concluded that the debtor countries' difficulties were not macroeconomic but arose instead through attempts to reconcile the competing interests of different interest groups within their own populations.

Stephany Griffith-Jones (Institute of Development Studies, Sussex) presented a paper on `Bargaining on Latin-American Debt: Theories, Practice and Policy Conclusions', which summarized the results of a large-scale research project. She posed three questions. How can one explain the nature of the debt deals which have made since 1982, which many observers agree have been closer to the interests of creditors than of debtors? To what extent have different countries been able to make different deals and how have the deals varied over time? What improvements to the bargaining process should be sought in the future? Griffith-Jones argued that there were several reasons why debtor countries have not pursued their own interests more aggressively in the past, but which may lead them to become more aggressive in the future. Debtors have been concerned about the possible effects of debt repudiation on the stability of the world financial system; but they may now be influenced by the reduced vulnerability of the banking system to sovereign debt defaults. Debtors have also been influenced, Griffith-Jones argued, by the perceived net benefits of an international system that provided them with substantial net resource inflows in the 1970s; this may now be changing as the inflows are matched or even exceeded by net outflows in the 1980s. Griffith-Jones pointed to internal factors in debtor countries (such as distributional issues or political change) that also affected their negotiating positions. She argued that a new approach to the debt problem is needed: it is in the interests of the debtor countries to develop credible macroeconomic and developmental strategies and to adopt tougher bargaining positions; in addition, differences of interest among creditor countries should be reflected in the negotiations.
In discussion of the Griffith-Jones paper, Kenneth Kletzer argued that it was inappropriate to make calculations of net resource transfers without using a discount rate. There was also discussion of whether banks' bargaining power was weakened when they make provision for bad debt.

Policy Conflict and Coordination
The workshop on 15 January 1988 was concerned with international policy conflict and coordination. Christopher Bliss (Nuffield College, Oxford, and CEPR) presented `A Model of Negotiation with Imperfect Information'. Bliss addressed two problems: the general issue of whether the modern formal theory of bargaining can be applied to the study of international negotiations; and the specific issue of analysing what it means to `delay agreement' or to `break off negotiations'. Bliss characterized the conventional analysis of bargaining, where the parties all have the same information, as implying that rational bargainers will always settle `out of court' rather than enter into long and costly negotiations; though the `out of court' settlement will depend on what would have emerged from the formal negotiations had they been entered into. He therefore argued that one function of international negotiations may be to elicit information from the participants. A delay or even breaking off of negotiations may be a rational strategy for eliciting information known to only one of the participants. Bliss concluded that the analysis he had been using faced two major difficulties: the participants are required to have very detailed information about other participants (in spite of the imperfection of the information) and to make extremely complex calculations; and there might be an incentive for participants to depart from their pre-announced commitments.
Lawrence Freedman (King's College, London) took the view that in the real world negotiations have many more dimensions than Bliss's model allowed: many different games are being played at the same time, and agreement is only one of many objectives. Freedman thought that discovering others' intentions may be much more important than determining objective facts: in disarmament negotiations, for example, there is usually little disagreement about the facts, but it may be necessary for one party to undertake a costly programme simply to inform the other party of the nature of its objectives.
John Cross (University of Michigan) outlined the recent work of a multidisciplinary group of social scientists at Michigan studying different aspects of international economic relations in a research programme funded by the MacArthur Foundation. Cross also presented a paper on `Obstacles to International Arms Agreements: Perspectives from Negotiation Theory', in which he analysed why arms negotiations seem generally not to have been successful, even though there seem to be mutual benefits from reductions in armaments. Cross suggested one explanation: the interests and objectives of politicians and negotiators may differ from those of the general public. Negotiators may be content with an outcome in which armament levels are high, and the costs of this outcome to the general public are spread so thinly across the general public that no one has sufficient incentive to involve themselves in the negotiating process. A second set of explanations offered by Cross focused on risk. At low armament levels even small differences in armament levels may have big effects. In addition, the risk of technical failure may be more serious at low armament levels. The cost of increased armaments could be justified as a means of reducing these risks. Cross also pointed out that the formal theory of bargaining suggests that the negotiated outcome is strongly influenced by the nature of what will happen in the absence of an agreement. Parties may therefore have a strong incentive to build up armaments simply to strengthen their negotiating positions. Cross concluded that the main hope for making arms negotiations more successful lies in changing the preferences of the negotiating parties, making them less concerned about the risks of small asymmetries in armaments and more conscious of the costs of armaments.
Andrew Hughes Hallett (University of Newcastle-upon-Tyne and CEPR) presented the final paper to the second workshop. In `Cooperation or Tariff Wars: Is US Protectionism a Threat to Economic Stability?', forthcoming as a CEPR Discussion Paper, Hughes Hallett employed a multi-country macroeconometric model to study the impact of import restrictions imposed by the United States as a means of reducing its trade deficit. Hughes Hallett investigated whether such a policy would increase US growth and thus US demand for the products of other countries, so as to outweigh the direct adverse effects on the other countries of US protection; whether it would be in other OECD countries' interests to retaliate; whether monetary and fiscal policies are more effective than tariffs for achieving trade balance; and whether international cooperation in trade policy is more important than in macroeconomic policy.

Hughes Hallett provided a numerical answer to these questions using the Federal Reserve Board's quarterly macroeconometric model linking the economies of the United States, Canada, West Germany, Japan and the United Kingdom. Taking a set of target values of various macroeconomic variables for each country and assuming that exchange rates are flexible, he computed the optimal macroeconomic and tariff policies for each country. His results indicated that, with flexible exchange rates, a macroeconomic policy package including tariffs is hardly more effective than one which excludes them, and that the optimal tariff levels are quite low. Other countries have little incentive to retaliate against US tariffs, but the negative impact of a tariff war on GNP is appreciable and unevenly spread between countries. Hughes Hallett compared these results with those obtained from a coordinated international policy package and found that coordination produced generally superior outcomes. Under coordination, other countries assisted US economic performance and agreed to fund&nbspa larger US budget deficit; in return they enjoyed lower inflation than in the equilibrium without coordination.
In response to a question about why tariffs were of any value under flexible exchange rates, Hughes Hallett observed that the countries in his model had three macroeconomic objectives, growth, inflation and the government deficit, but only two policy instruments, the money supply and government expenditure. Rob Eastwood (University of Sussex) sug- gested that the introduction of more fiscal policy instruments would remove any incentive to use tariffs.

East-West Economic Relations
The third workshop of the series focused on `East-West Economic Relations'. Barry Buzan (University of Warwick) and Gautam Sen (LSE) presented a paper on `The Economic Impact of the Arms Race', which assessed the effects of military R&D expenditure on the development of civil technology. Buzan and Sen identified the fundamentally anarchic nature of international political relations as the driving force behind military technological innovation, but observed that it was not until the late nineteenth century that military technology began to assume the `leading-edge' position that it now holds. The twentieth century has seen both increasing state involvement in technological innovation and growing divergence of the needs of civil and military technology. The authors discussed four major civil industries whose postwar development has been fundamentally influenced by the dynamics of military R&D priorities: nuclear power, space satellites, civil aviation and computers. In all four cases, though to different degrees, the development of the civil industry has been very heavily influenced or even driven by military needs. The most important effect of military R&D, Buzan and Sen argued, was the `early' or `premature' introduction of certain industries and the decline of competing industries; but they took no general normative position on whether the net effect was beneficial or harmful.
In discussion, Ron Smith (Birkbeck College, London) argued that the standards of reliability acceptable in military applications were very much lower than in civil applications and that the successful diffusion of technology to the civil sector typically occurred only when production was not in the hands of military contractors. As evidence he cited Boeing's separate civil aviation division and IBM's origins outside the military sector.
The second paper, `Benefits Beyond Belief: Assessing the Cost of US Export Controls', was given by Stuart MacDonald (University of Queensland). MacDonald argued that US controls on exports to Eastern bloc countries have been justified in more than one way: not only do they fulfil the foreign policy role of expressing views about Soviet bloc policies, but they also prevent advanced technology reaching the Soviet bloc. MacDonald also argued that there is a tendency to ignore the social and economic costs of these export controls and that this tendency has grown with the ascendancy in US policy-making of the Department of Defense over the Commerce and State Departments. A critical point came in 1976 with the Bucy Report, which argued that controls on information were the most important aspect of export control. One consequence has been US attempts to control academic research in sensitive areas and to control unclassified information in government telecommunications and computer systems. MacDonald contrasted two 1985 estimates of the economic effects of US export controls. A Defense Department report estimated the benefits to the United States as $13 billion over a 13 year period, the result of savings on defence expenditure which would have been required if the Soviet bloc had obtained the controlled exports, while a CIA report estimated the costs to the United States of export controls at $9.3 billion annually. MacDonald believed that the latter calculation was more soundly based. He also argued that the celebrated case involving Toshiba reflected a battle for influence between the Departments of Defense and of Commerce rather than real issues of national security. He concluded that a system of ineffective export controls, manipulated for domestic political advantage, illegally and arbitrarily enforced, seems guaranteed to engender discord, distrust and ultimately contempt.
In his comments on MacDonald's paper, Richard Portes (CEPR and Birkbeck College, London) called attention to a substantial economics literature establishing a strong case against all but the most specifically military controls. He argued, however, that the existence of `turf' battles between US government departments was neither news nor evidence for or against export controls, and he queried the reliability of any cost-benefit analysis of export controls. Barry Buzan thought that MacDonald's study was interesting as a study of bureaucratic battles, but that the numbers themselves were of little significance.

The third paper at this workshop examined `The Strategy of American Economic Diplomacy Towards the Soviet Union in a Period of Transition: The Nixon-Ford Years'. John Cable (Hertford College, Oxford) argued that the United States has consistently relied on economic leverage to influence Soviet behaviour during the past two decades, even though the form this leverage has taken has varied from comprehensive economic denial to `positive linkage' under Nixon and Kissinger, `negative linkage' under Carter, and `squeezing' the Soviet Union under Reagan. The core of the Kissinger policy, Cable stated, was to contain the risks of conflict by developing a series of&nbspfunctional bilateral arrangements with the Soviet Union. US foreign policy was to be geared to concrete national interests rather than to abstract notions of ideology. The US government did not harbour the illusion that instruments of trade and finance alone could create a harmonious international order, but it hoped that relaxing political barriers to East-West commerce, liberalizing the CoCom strategic export controls, and extending credits on favourable terms would give the Soviet leadership an incentive to pursue a policy of peaceful coexistence with the rest of the world.
Comments on Cable's paper focused on the difficulties confronting a US administration which attempts to develop a finely tuned set of foreign policy signals in economic affairs when it cannot control the behaviour of Congress or of private firms and when it is subject to pressure groups such as the farm lobby.

Conflicts over Natural Resources and Agriculture
The concluding workshop took as its theme `International Conflicts over Resources and Agriculture'. Alexander Sarris (University of Athens) presented a paper on `Food Security and International Security'. He quoted recent World Bank estimates that one billion people suffer from food insecurity, in the sense of having fewer calories than the required intake for normal activity. Such people are particularly vulnerable to the effects of market fluctuations. In less developed countries, there exist insurance mechanisms such as crop diversification, labour market participation, and asset accumulation. Poor countries seeking food security through foreign trade face difficulties, Sarris argued, as do poor individuals seeking security through markets. Prices tend to be high when they need to buy and low when they wish to sell. Sarris presented a formal model of public and private stockholding that could be used to analyse the optimal response to supply fluctuations. The model indicated, according to Sarris, that world food security required the holding of substantially higher levels of food stocks than are currently held. Developed countries' agricultural policies not only reduce the prices received by developing country exporters, but also increase the variability of prices in world markets. Sarris argued that food security considerations could lead developing countries to react to increased price variability with increased production, where static efficiency considerations would have suggested the opposite response. He advocated the development of improved multilateral international insurance mechanisms as the best response to food security problems.
Stephany Griffith-Jones welcomed Sarris's analysis, which went beyond the usual treatment of food security as a purely technical issue. She also noted the strong elements of conditionality in many food aid schemes. Several participants pointed out that there are more complex interactions of interest groups than a simple North-South division would suggest. Interest groups such as urban elites in developing countries and farm lobbies in developed countries promote developing country policies which discourage cash crops. Ron Smith noted the connections between food and internal security: from the viewpoint of national elites, food emergencies may be politically helpful, and arms expenditure may be better value than food.
In his presentation to the workshop, L Alan Winters (University of Wales, Bangor, and CEPR) addressed `The National Security Argument for Agricultural Protection' (a revised version is now available as Discussion Paper No. 287). Winters cited Japan, the UK, Switzerland and Sweden as examples where national security had been officially used as a justification for agricultural protectionism. He focused not on the question of whether the national security argument was valid, but whether current policies are a cost-effective means of achieving national security objectives. He discussed the economics of preparing for an import embargo and argued that official intervention is required before the embargo only if there is some link between pre- and post-embargo production or consumption and if the private sector fails to take proper account of the prospective embargo in its planning. Stockpiling is the most important example of an action taken by the private sector in anticipation of an embargo. Winters next argued, on the basis of both econometric demand studies and studies of nutritional requirements, that food is not uniquely vulnerable to an embargo and that even a fully effective embargo on British food imports would impose costs which were tolerable. An effective embargo was unlikely, however: agricultural products are difficult to embargo successfully because of the range of sources of supply. On the other hand, a successful oil and chemical embargo is more likely and British agriculture depends heavily on these inputs, so the apparent self-sufficiency is `skin deep'. Winters concluded that increased levels of stocks, both of inputs and of agricultural products, would be a much more effective means of meeting security objectives than the promotion of national self-sufficiency in food production. He noted that the Soviet Union and other East European countries were the greatest beneficiaries of the OECD countries' excessive agricultural protection, and he quoted one estimate that these policies had resulted in a transfer of $23 billion from West to East in the early 1980s.
Participants found Winters's arguments convincing. There was scepticism as to whether governments really take the national security argument seriously, and the point was made that, in the British case, postwar agricultural policy was made not so much to meet national security objectives as to pay back wartime friends, such as the farm sector and New Zealand.
The final paper, `Sustainable Natural Resource Management as a Factor in International Economic Security', was given by Edward Barbier (International Institute for Environment and Development, London). He argued for the importance of environmental factors in international economic security. Environmental degradation in less developed countries may leave their economic systems more vulnerable to external shocks and stresses. These countries are particularly vulnerable to fragile ecological processes, and there may be a tendency for cumulative degradation to take place, as poor people are driven by economic circumstances to damage their environment in ways that increase their poverty. By the year 2000, the number of agricultural households which are subsistence farmers, pastoralists or landless will increase by 50 million to nearly 220 million. These resource-poor households will be caught in a poverty trap that induces them to over-exploit existing resources. The stresses placed by increasing environmental degradation on the economic system may have two scarcity effects: a short-run decline in essential health and environmental services, and a longer-run collapse of natural and managed eco-systems. Barbier concluded that the importance of sustainable resource management to overall economic security and welfare should merit its inclusion as a major development objective.
Commenting on Barbier's paper, Ron Smith noted an external security dimension: the possibility of external coercion to stop international environmental degradation. There was also discussion of the extent to which improving environmental conditions and increasing trade may increase countries' economic vulnerability.