Portal de Periódicos da CAPES

Economic contagion and a pro‐poor social epidemiology

2020; Wiley; Volume: 52; Issue: 2 Linguagem: Inglês

10.1111/josp.12375

1467-9833

Darrel Moellendorf,

Employment and Welfare Studies

The world built by financial integration allows crises that move quickly from one country to another, destroying the lives of people who are vulnerable to poverty. Witness the remarkable speed of the crisis that began as a home mortgage crisis in the United States in 2007 but quickly generalized into an economic crisis around the world. When such contagion breaks out, the poor typically suffer disproportionately. Even after a recovery, the lasting effects, the hysteresis, can trap an entire generation in poverty, which they might otherwise have avoided.1 The poverty produced by means of economic contagion is plausibly an example of what Iris Marion Young calls a structural injustice. Contagion is a powerful metaphor. One worthy theoretical project of global justice would be a pro-poor social epidemiology of this contagion. Such a pro-poor epidemiology would have many elements. In this paper, I limit my efforts to just two: Explaining the key role of crisis transition mechanisms in producing contagion; and in light of that, discussing how best to understand responsibility for reforming the institutions of the global economy in order to remove or mitigate the effects of crisis transition mechanisms. Although the contagion that spreads poverty can be understood, following Young, as a structural injustice, her social connection model of responsibility is less adequate for understanding responsibility in this case. To be vulnerable to poverty as a result of an economic crisis is to stand in a relationship of susceptibility to a catastrophic loss of the means of economic security due to forces beyond one’s control. The general idea is that vulnerability is a relational property. For any particular person (p), she is vulnerable (V) to some bad outcome or happening (o) due to various factors (f1, f2, f3, etc.). Vulnerability can be decreased or eliminated by either reducing or removing the power of one or more of the factors that affect the person or by empowering her to protect herself against the factors.2 The standardly accepted rights of ownership and exchange are not the authoritarian extravaganzas of a heartless Nero or some brutal Genghis Kahn. They are, rather, parts of the standard legal rules of ownership and exchange that govern people’s lives in much of the world. But when they are not supplemented by other rights (e.g., social security, unemployment insurance, public health provisions), these standard rights may operate in a way that offers no chance of survival to potential famine victims.3 People are vulnerable to starvation because they have no entitlement to food when market forces render it too expensive. Identifying the factors causing vulnerability is important for a remedy. Remedying vulnerability to starvation could include establishing legal entitlements to sufficient food for healthy survival or providing supplemental income. Crisis transition mechanisms cause the contagion of economic crises, and are a major factor in vulnerability to severe impoverishment. In the aftermath of the home mortgage crisis in the United States, three important pipelines of cash to the developing world slowed to a trickle, official capital inflows, private capital investments, and private remittances. In 2009, net inflows of private capital were down by nearly 70% of their 2007 peak.4 Low-income people suffered as their already tiny share of the capital inflow dwindled to next to nothing. The value of shares on several equity markets plummeted. In Latin America and Asia, the average rate of decline was around 50% by the spring of 2009. Also, as the cash inflows slowed, the cost of borrowing increased five- to sixfold.5 Although more stable than capital inflows, remittances also dropped by about 6% in 2009.6 Remittances have become a very important source of income for many developing and low-income countries, amounting to more than three times the official development assistance. For many developing countries, they are the largest source of external income. Overall, the contagion produced a precipitous and crippling fall in economic activity. World economic output fell from growth rate of 3.9% in 2007 to 3.0% in 2008 to −2.2% in 2009, the first time in 60 years that the global economy had contracted.7 Emerging and developing countries saw growth rates fall from 8.3% in 2007 to 8.1% in 2008 to 1.2% in 2009.8 The fall in growth in this group of countries is especially significant for poverty eradication efforts. The World Bank estimates that for every percentage point in growth lost 20 million people are trapped in poverty.9 Although the effects were grim for global poverty, they were not as bad as the possibility suggested by that estimate. According to the Bank’s report in 2010, “Some 64 million more people around the world are expected to be living on less than $1.25 a day by the end of 2010 than would have been the case without the crisis, and between 30,000 and 50,000 children may have died of malnutrition in 2009 in Sub-Saharan Africa because of the crisis.”10 The language of “mechanism” suggests a reliable causal relation according to which an outcome can be explained.11 As mentioned above, identifying a factor of vulnerability is important for minimizing or removing vulnerability; mechanisms can sometimes be neutralized. As an example, consider the way in which Drèze and Sen explain famines by the mechanism of the entitlement system. “If people go hungry on a regular basis all the time, or seasonally, the explanations of that have to be sought in the way the entitlement system in operation fails to give the persons involved adequate means of securing enough food.”12 According Drèze and Sen’s account, the mechanism by which famines come to ravage populations is roughly that food distribution systems based wholly on the market, with no entitlements other than to the food that one can purchase, do not secure entitlements for a large number of people in times of food shortage, rendering them vulnerable to famine; during shortages people lacking of such entitlements suffer famines. In short, famines arise due to a failure to establish effective entitlements other than property rights in the foodstuffs that one purchases. Crisis transition mechanisms explain contagion in roughly the following way: Economic down turns in highly developed countries reduce capital investments in developing countries, and they reduce demand for products produced in both developing and less-developed countries as well as the remittances sent to both classes of countries; the cumulative effect of reduced capital investments, demand for export products, and remittances received is diminished economic growth in such countries. Factors causing vulnerability to famines according to Drèze and Sen’s account include poverty, a market distribution system for foodstuffs, and a lack of additional entitlements to food. One has an entitlement just insofar as one qualifies under a legal system with certain recognized property rights. Being the bearer of the entitlement justifies one in demanding that it be honored. People are vulnerable to famines in times of shortage if they lack food entitlements that could be provided by a well-functioning social and legal system. Vulnerability is to famine is a condition of people within a system of social and legal entitlements; it is socially constructed. Being vulnerable to poverty in times of economic crises bears some similarities. The factors producing vulnerability include living in a country susceptible to contagion in virtue of the crisis transition mechanisms and being entitled to income mostly or only through a stream that is conditioned directly or indirectly on foreign investments, exports, or remittances. Such an income stream is governed by a system of property rights. Hence, the legal entitlement to income depends on the property rights that are recognized. Vulnerability to poverty is the result of occupying a socially constructed role—the bearer of an entitlement—in which one is only precariously protected against loss of income. Remedies could affect the existence of such roles either by protecting income streams or by supplementing them with other streams that are unconditional. There is something perplexing about the forgoing explanations. Vulnerability in both the case of famine and poverty is due to an absence, a policy failure. In the case that interests us in this paper, poverty caused by the transmission of an economic crisis, there is a failure to regulate or insure against crisis transition mechanisms. At first, gloss that sounds plausible enough. But how can absence cause anything at all, let alone poverty? Causes, presumably, are real. How, then, can something non-existent be causally efficacious? There must be more to the story of crisis transition mechanisms. First back to the case of famines, analyzed by Drèze and Sen as a failure to establish food entitlements. For the reason just highlighted, it would be implausible to attribute the famine merely to not acting, in this case a failure to legislate or to otherwise ensure entitlements. For that would be to attribute causal efficacy to nothing, a non-action. More troublesome still, it would explode the class of causes. For every action taken, there is a potentially infinite set of possible actions foregone. It cannot be the case that all of these failures to act are causes. Still, even though it is imprecise, talk of failures producing outcomes is familiar to us. Consider this charge of a tort: The mechanic’s failure to tighten the bolts resulted in the wheel falling off, causing the accident. We all understand what that charge means. It is a counterfactual claim. If she had tightened the bolts, the wheel would not have flown off. Drèze and Sen are also making a counterfactual claim; if there had been a system of entitlements to foodstuffs, people would not have gone hungry on a mass scale. We might, however, ask why the mechanic’s failure to tighten the bolts is any more responsible for the outcome than the failure of the dishwasher in the restaurant next door? In both cases, if they had tightened the bolts, the wheel would not have flown off. Consider the attribution of outcome responsibility in tort law as explained by Tony Honoré. According to Honoré, attributing outcome responsibility to a person is not necessarily assigning causal efficacy to her, rather it is to credit or discredit her for an action or omission.13 The point is not whether an agent caused an outcome. Rather, it is whether she can be credited or discredited for the outcome. Now, crediting and discrediting suggests some kind of normative appraisal. There is a seemingly infinite number of actions not performed; outcome responsibility cannot be attributed to every one of them. Not all of them are legal failures. Outcome responsibility can only reasonably be attributed to some very small subset of non-actions. In the case of the law, this subset is often characterized by failures to act in accordance with a valid legal norm.14 An outcome can be credited to a person if she fails to act in accordance with a valid legal duty. The mechanic is outcome responsible for the tire flying off, but the dishwasher in the restaurant next door is not, because the mechanic’s failure to tighten the bolts, unlike the dishwasher’s, contravenes a valid legal norm. Suppose one used this method of attribution of legal failures in the case of famine attribution. In that case, the claim that legislative failures to establish entitlements are the mechanisms by which famines are produced, would be understood as a claim that famines are credited to legislative inaction. But to be credited to legislative inaction, rather than the infinite number of other non-actions, it must be the case that the legislature had a valid duty to act. It is not necessarily a legal duty, however, since there might be no legal duty of the legislature to ensure such entitlements. In this case, the explanation of why the legislature failure is the cause of the famine would seem to rely on the moral claim that the legislature should have established entitlements to food. Attributing legislative inaction as outcome responsible on the model of legal attribution is problematic for two reasons. First, it would render the Drèze and Sen account normative. Although they might have been thought to be doing positive social science, attempting to understand the causes of famines, as the first step to remedying the problem, in fact the explanation itself would be normative; it only makes sense if there is moral duty of legislatures to act. The language of “mechanism” sounds explanatory, but in fact it would be normative. This is closely connected to the second problem, which is that is seems to get things backwards. Positive social science seems to have an important independent role to play in correcting social ills such as famines. If we can understand how famines come to pass, then we can understand how they might be prevented. After understanding that, we can develop a moral account of who is to be held morally responsible when constructing alternative systems of entitlements. But the legal attribution model requires that we already have the answer to at least one question of moral responsibility. In light of these problems, let's consider an alternative approach to attributing an outcome to a policy or institutional failure. Instead of taking the outcome to be caused by some morally relevant non-action picked out from an infinite set of non-actions, take it to be the effect of other real policies. The identification of the factor is non-normative. This seems to fit better with Drèze and Sen’s approach. A system of property rights in which the market plays the sole role of distributing foodstuffs is that factor the produces a loss of effective demand for many people when prices are driven up. But relying only on market forces is a policy choice. For that reason, one can credibly claim that policies have caused the famine. We understand the cause to be the actual policies, not a lack of policies. The explanation rests on causal rather than normative attribution; the mechanism is positive, not normative. The preservation of the causal relations all of the way down is the chief advantage of the second approach over the first approach. Impoverishment caused by the contagion of economic crises seems to be a good example of what Iris Marion Young calls a “structural injustice.” For in cases of structural injustice, “The actions of particular persons do not contribute to injustice for other persons directly…but rather indirectly, collectively, and cumulatively through the production of structural constraints on the actions of many and privileged opportunities for some.”15 A person who purchased a home in the United States and refinanced that purchase with a larger second short term mortgage in order to pay off some acquired consumer debt, but who found herself unable either to pay for the remainder of the amount owed on the house or to refinance again, did not directly cause the impoverishment that came to be in the developing and less developed world. Nor did the real estate agent who facilitated the house purchase, the mortgage broker who arranged the loan package, the bank employee who approved the loan, the other bank employee who sold the loan bundled with others in order to manage the risk, the employee of the financial corporation who purchased the bundle of loans, and so on. But it seems reasonable to say that all of them contributed to the crisis indirectly. The injustice is structural in the sense that it is not attributed to any particular person’s action but to a set of actions permitted or encouraged by the structure of existing institutions. As Young notes, both the law and the ethics of responsibility often look backward. In order to attribute backward looking responsibility it is important to identify the agent who brought about the blameworthy event or should have acted to prevent it. The backward looking conception of responsibility that employs these ideas, Young calls “the liability model.” She notes that the liability model seems appropriate “for the purposes of sanctioning, punishing, or exacting compensation or redress.”16 Typically, but not always, one must have been at fault to be held morally or legally liable for an accident. Fault requires a culpable failure to act according to a duty. The law usually recognizes a small class of exceptions to fault known as strict liability. For actions in this class, one is held responsible despite doing nothing for which one could be faulted. Strict liability remains somewhat controversial morally, but typically it is defended in the case of particularly dangerous or important classes of activity and when one is put on prior notice that by acting one is assuming liability for an action.17 The relevant point, however, is that dominant idea of the liability model is that an outcome of harm to persons or of damage to property can be attributed to an agent’s action or culpable inaction. The person refinancing the home, the mortgage broker, and the bank employee are not to be blamed for the global recession, even if they participated in the structural injustice of the economic contagion. Young rightly rejects the liability model of responsibility for structural injustices. In such cases liability cannot be attributed to agents because “it is not possible to identify how the action of one particular individual … has directly produced harm to other specific individuals.”18 But, agents are involved in the reproduction of social and political structures and their actions permitted or encouraged by these institutions may have unjust outcomes. Young believes that this activity of reproduction by individuals is relevant to an account of responsibility, even if it does not entail liability and blame. For cases of structural injustice, she defends what she calls “the social connection model” of responsibility, according to which “all those who contribute by their actions to structural processes with some unjust outcome share responsibility for the injustice.”19 (Emphasis added.) Liability model: An agent is responsible for an outcome only if she is outcome responsible for it. Social connection model: An agent is responsible for a structural process producing unjust outcomes if her action contributes to the structural processes. Being outcome responsible requires either causing the outcome or not preventing it when it was in one’s power do so. The contrast, contributing to a structural process, needs to be clarified. Young explicitly rejects any requirement of directly producing the injustice. Her examples of contributors include both consumers who create demand for products made under unjust work conditions and producers who make products in those conditions. Still, a clear account of what it is to contribute to an injustice is required in order to evaluate the account. Causation, one might think, is clearer than contribution. In fact, however, controversy abounds in jurisprudence about how to attribute causation, and ethics is beset by the same problems. One common way to think about causing harm is but-for causation. X causes Y only if X is a necessary condition of Y. That, however, is often an unsatisfactory account of causation since it does not identify causal factors in cases of over-determination, where more than one factor could have been sufficient alone to cause the outcome. If either of two penetrating bullets shot by two different people would have killed someone without the other, then according to but-for causation neither shooter caused of the death. More sophisticated versions of causation require a cause to be a necessary element of a number of events that together where sufficient to produce the outcome. More exactly, X is a cause of Y only if X is a necessary element of a set of conditions that are sufficient for Y.20 Throwing a rock, which broke the car window, is the cause of the subsequent accident, even if the accident also required the driver to brake improperly on an icy road, and if the accident might have occurred without rock throwing due to speed at which the car was going on the icy road. According to this account of causation, we might hold the thrower responsible. But in the case of structural injustices, an individual action—any particular default on a mortgage, for example—is not even necessary element of a set of sufficient conditions. The collapse of the market did not require any particular person defaulting. When a broker sells a home mortgage in the United States, she does not cause a drop in the demand of basic commodities in less-developed countries or a reduction in the remittances sent to persons in the developing countries. The sale and purchase of the mortgage, along with the sale and purchase of a great many others, creates a real-estate bubble. When the bubble bursts the result is large scale defaulting on loans, which, depending on how the risk is bought and sold, can spread from the financial sector to the real economy and then via crisis transmission mechanisms across borders. The harm occurs only because of the actions of many other people and only in virtue of the rules of various institutions, including the real estate market, the home mortgage market, the banking sector, the market for various financial instruments, and so on. If any particular person decided to quit selling home mortgages, there would be no change in the harmful outcome. Although as a matter of fact there seems to be evidence of liability responsibility for the financial crisis, even despite a paucity of prosecution attempts, none of the actions discussed here would render an individual liability responsible.21 In order to assess Young’s account, we have to get clear on what it is for an agent to contribute to a structural process. Doing so may be just as tricky as being clear on what it is for an agent to cause an outcome. Young takes the class of those persons who contribute to structural process causing injustice to be much bigger than the class of those causing the injustice since the latter is, she thinks, implausibly small and the former includes even those who suffer injustice. One proposal is to include in the class of contributors all those people who follow the rules, either formal or informal, of the institutions comprising that process generating the injustice. The action required by the rules may be unobjectionable, such as refinancing a house, but the rules nonetheless constitute institutional structures that produce injustice. According to this proposal, contributors are rule followers. In the case of the financial crisis, contributors would include not only bank managers and stockholders, but also mortgage brokers, consumers of home loans, consumers of basic commodity imports, as well as producers of basic commodities in poor countries and recipients of remittances. This proposal conforms to Young’s view that the contributor/non-contributor distinction does not track the oppressor/oppressed distinction. For even the workers in the sweatshop are contributors. In the context of assigning responsibility for the financial crisis, however, the proposal that contributors are institutional rule followers is problematic. The crisis unfolded in part because millions of people defaulted on their home loans. Presumably, the social connection model should identify them as among those who contributed to the crisis. If not them, then whom? But those defaulting did not follow the rules. On the contrary, they violated the terms of their contractual obligations, and in doing seemly contributed. Hence, the proposal that contributors are limited to rule followers limits the class of contributors too narrowly to capture everyone who seems to have contributed to the financial crisis. Consider an alternative proposal that contributors include not only institutional rule followers, but anyone responding to her role as assigned by institutional rules, regardless of whether she adheres to the norms of the role. Accoridng to this second proposal, contributors are rule responders. The proposal has the virtue of including people who defaulted on their loans since defaulting is a way of responding to a lending contract. Rule breaking is a way of responding to a rule, but sometimes such responses are done deliberately to change the rule. That suggests a problem. The worker on strike for better pay and work conditions, or the consumer who refuses to buy products made in highly exploitative conditions, are responding to institutional rules, and it would seem strange to say that the activity of striking and boycotting makes the employees and consumers responsible for the injustice. Presumably, what the account should conclude is that these sorts of actions are appropriate in light of the participation-generated responsibility of employees and consumers. Young’s examples of contributors tend to focus on people who follow the rules. She suggests that contributions can include “actions and policies that most people consider normal and acceptable, or even necessary and good,”22 that agents often possess “little awareness of how their actions contribute” to injustices,23 and that “it is in the nature of such structural processes that their potential harmful effects cannot be traced directly to any particular contributor to the process.”24 Any action in response to one’s role assignment, then, seems too broad a category for contribution. However, we have seen that the class of contributors to structural injustice needs to be broader than mere rule followers, to capture the defaulters. Nonetheless, the class of contributors should not include everyone acting in response to a role since efforts to change structural injustice should not render one responsible for it. The person defaulting on her mortgage is not typically seeking to change the rules. She is rather seeking relief from her obligations. It would be desirable for an account of contribution to include the person defaulting but not those breaking the rules by resisting oppression. They key is to carve out an exception for those who in responding to their roles seek to change the rules. Consider then a third proposal. Contribution to structural injustice includes all action in relation to institutional role assignment, except that which seeks to change the way institutions assign outcomes. Those seeking only relief for themselves—home owners defaulting on loans—are contributing to structural injustice, while those resisting rules—workers on strike—are not. According to this third proposal, contributors are rule responders, except those intending to change the rules. Although there are controversies about how best to understand causation on the liability model, the model survives because there are morally important reasons to think that a person who causes an injustice should be held responsible. As Young recognizes, holding people liable for their actions is a way of recognizing and respecting their agency.25 It assumes that people are able to control their actions in light of the foreseeable consequences, and that they can be influenced by reasons to do so. But why hold people responsible for contributing to institutions in the absence of causing outcomes? Why hold people responsible either for acting in ways assigned to them by institutional rules or for seeking relief from the strictures of the rules, especially when their actions seem reasonable given the alternatives? We are not expressing respect for someone’s agency when it is not the outcomes of their action, but their contribution, that matters. What moral purpose then is served by such an account of responsibility? And what are the moral costs of doing so? Consider the consumer who defaults on her home loan. She is unwittingly contributing to a set of market responses that have powerful effects on the well-being of people in poor countries. Someone might object that to hold her responsible for such actions in this case is to demand too much of her. She cannot be expected to continue to make loan payments if she has to choose between doing that and buying food and clothes for her children. Young’s account can deflect the immediate force of the objection. Holding a person responsible under the social connection model is not for the purpose of sanctioning her for her action. Instead, responsibility claims according to the social connection model are forward looking.26 “Being responsible in relation to structural injustice means that one has an obligation to join with others who share that responsibility in order to transform the structural process to make their outcomes less unjust.”27 One need not be outcome responsible to be responsible for changing an unjust social structure. When the architects of our republic wrote the magnificent words of the Constitution and the Declaration of Independence, they were signing a promissory note to which every American was to fall heir. This note was the promise that all men, yes, black men as well as white men, would be guaranteed the unalienable rights of life, liberty, and the pursuit of happiness. It is obvious today that America has defaulted on that promissory note in so far as her citizens of color are concerned.28 If the concept of responsibility involves debt, one way to think about the differences among conceptions of responsibility is that they vary according how the debt is incurred and what must be done to relieve it. King believed that the responsibility was general; every American was heir to the responsibility to establish racial equality. It was not a special responsibility of a few. Conceptions of responsibility also vary by purpose. It makes little sense to evaluate an account without understanding its purpose. The liability