Abhijit V. Banerjee and Esther Duflo, Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty, PublicAffairs, 2011; pbk £18.99 or less.
|
RATING: 24
|
Buy this book?
|
No
|
The Nobel Prize for economics, or the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel, to give it its full title, was awarded in 2019 to Abhijit Banerjee, Esther Duflo and Michael Kremer, for their experimental approach to alleviating poverty. It is an opportune time, therefore, to look again at Banerjee and Duflo's Poor Economics, published in 2011. For avoidance of doubt, I should make clear that nothing I say about it has any implications for the work of Michael Kremer, which is generally judicious and level-headed; and his recent jointly-authored survey of behavioural developmental economics (Kremer et al, 2019) is a good guide to the field.
A distinction also needs to be made between randomised controlled trials (RCTs) on the one hand, and Poor Economics on the other. There is a substantial critical literature on RCTs - 'lab' or field 'experiments' modelled on the method by which new drugs are tested, through which the prize recipients have made their reputations - some of it predating the publication of Poor Economics (notably, Heckman and Smith, 1995; Ravallion, 2009; Rodrik, 2009; Barrett and Carter, 2010; Cartwright, 2010), some following its publication (notably Ravallion, 2012, Reddy, 2012, Deaton and Cartwright, 2018, Ravallion 2018), and some following the award of the prize (notably, the dossier of short responses in World Development, 127, 2020, among which the persistent Ravallion, 2020). Together and independently, the prize recipients have also contributed to the debate (e.g. Duflo and Kremer, 2008; Banerjee and Duflo, 2009; Banerjee, Duflo and Kremer, 2019). The consensus of this literature (not shared by the most enthusiastic) is that RCTs represent a useful addition to techniques of evaluation, and that they work better with specific and delimited issues than with broad policy initiatives, but that they do not either escape all the problems that other methods face, or lend themselves unproblematically to scaling up, transfer to different locations, or estimates of long-term consequences. They do not resolve definitively issues of causality or efficacy, and there is a general concern that they are more beset by ethical issues than their counterparts in the medical world, but far less advanced in the development of protocols to address them.
All this, however, is largely tangential to Poor Economics, which is not only about RCTs, and in which none of the critiques of RCTs available when it was written is mentioned, let alone addressed. It is not a contribution to the debate, or even a scholarly exposition of BanerDu's approach. Rather, it is a piece of shameless self-promotion which gives precious little credit to the work of others, and which, despite the claim of 'radical rethinking', follows an entirely orthodox line of analysis derived from the World Bank of the 1990s. Far from being scientific in any sense, it is ideological to the core. Perhaps out of a sense of professional decorum, even the reviews of the book itself cited above (Ravallion, 2012; Reddy, 2012) don't tell you just what an appalling piece of hagiography it is.
From the start, its true heroes are not RCTs, or the extreme poor, but Banerjee and Duflo (hereafter BanerDu) themselves. Poor Economics invokes Esther and Abhijit, aged six, on the very first page (along with Mother Teresa!), hinting at a lifelong and indeed a saintly vocation. It then launches into a diatribe that accuses the development economics and policy professions at large of reducing the poor to cartoon characters, with no names given, or examples provided. 'Because the poor possess very little,' they say, 'it is assumed that there is nothing interesting about their economic existence' (viii). By whom? And what could this possibly mean? This is writing of the most light-minded kind, without any discernible thought process attached, and it sets the tone. BanerDu claim, in contrast to their peers, to have always taken 'the time to really understand [the] lives [of the poor], in all their complexity and richness'; they have been to the 'back alleys and villages where the poor live', and unlike those unspecified other '(often Western or Western-trained) professional development economists and policy makers', with the 'simple models' they have traditionally used to think about the lives of the poor, they have struggled through initial confusion to 'knit together a coherent story of how poor people live their lives' (viii-ix). In fact they use a simple model themselves (the S-curve, which addresses the possibility of poverty traps), and their ethnography, such as it is, is shallow and anecdotal; and Banerjee (Ph.D. Harvard, 1988, subsequently at Princeton, Harvard and MIT) and Duflo (Ph.D. MIT, 1999, and there ever since) are quintessentially mainstream development economists themselves. 'What is striking,' they say all the same of people on less than a dollar a day, in another witless pronouncement, 'is that even people who are that poor are just like the rest of us in almost every way' (ix), so managing not only to patronise them but also to dismiss entirely the significance of class difference. You know right away that you are in for an analysis that will be entirely uncritical of a system in which a small minority own the means of production, and the great majority are dependent on their capacity to work to survive. It quickly becomes clear, too, that they do not simply want to understand the poor. They want to change them. In a more accurate description of their purpose they say that this is 'a book about the kinds of theories that help us make sense of both what the poor are able to achieve, and where and for what reason they need a push' (x). And in the closing lines of the foreword, they give an early indication of the kind of push they need (one that will be familiar to anyone acquainted with World Bank policy):
'The book ... tells a lot about where hope lies: why token subsidies might have more than token effects; how to market better insurance; why less may be more in education; why good jobs matter for growth' (xi).
Everything, it turns out, is about making the global labour market work better.
The first step BanerDu take in embarking on their journey is to sideline what they call 'the "big questions": What is the ultimate cause of poverty? How much faith should we place in free markets? Is democracy good for the poor? Does foreign aid have a role to play? And so on' (3). They discuss only the last (and smallest) of these, and then only to say that there is no simple answer. But don't be deceived. This position on 'big questions' is by no means as naive as it looks, as they only affect agnosticism. Taking the first two in reverse order, they place their faith uncritically, not in entirely free markets, but in the private enterprise system, suitably regulated. Specifically, they endorse the rule of capital, and seek to turn the extremely poor into the productive poor, in part and with increasing reservations through petty entrepreneurship, but primarily, though late on and without extended discussion, through the provision of 'good jobs'. By affecting to eschew 'big questions' they bracket this preference off from debate. This leaves them by default with a clear and clearly inadequate position on the cause of poverty - it arises from mistaken policies on the part of well-intentioned governments, and poor choices by the poor, who act rationally within the limits of their own understanding (though with exceptions common to all of us, such as present bias and procrastination, which can be bracketed off in turn as 'human nature'), but lack the knowledge and resources to act in their own best long term interest as BanerDu perceive it.
All this means that there is very little that is new in their policy preferences. They are entirely in tune with such former Nobel prizewinners as Amartya Sen (1998), Joseph Stiglitz (2001) and Richard 'Nudge' Thaler (2017), and hand-in-glove with the leading institutions that manage the global capitalist economy. Their agenda echoes that of the World Bank point for point, and is equally oriented towards reforms that change behaviour by altering incentive structures. The very first World Development Report, published coincidentally in 1978, when Duflo was six years old and fixated on Mother Teresa, called for action on 'the structure of economic incentives, the allocation of investments, and the creation of special institutions and programs to increase the productivity of the poor and their opportunities for employment’ (World Bank, 1978: 65). In 1980 the Bank expanded in detail on nutrition, health, education and population control as key components of a strategy to increase the productivity of the poor (World Bank, 1980: 46-70). This is the water in which BanerDu have been swimming since they first dived into development studies, and always with the current. So much so, in fact, that the first four chapters of Poor Economics cover in turn nutrition, health, education, and population control - a full house, albeit in a four-card hand. Nor are the specific problems and potential solutions they take up new. So, for example, the first focuses on the efficacy of readily available and inexpensive nutritional supplements, such as iron and iodine. And although they repeatedly claim that (again unspecified) development economists and agencies continue to neglect such micro-nutrients, arguing instead that the problem is a deficiency of protein and calories, these precise micro-nutritional deficiencies - iron and iodine - were highlighted in the 1980 World Development Report itself (p. 63), and have been the focus of attention ever since. BanerDu must be well aware of this, just as they show themselves to be aware that periodic food 'shortages' do occur, arising from combinations of crop failure and maldistribution, and do lead to serious famines from time to time. Mystifyingly, they remark that: 'The importance of micronutrients was not fully understood, even by scientists, until relatively recently' (34). I'm not so sure about that. The health-related properties of iodised salt were identified in the 1830s; medical RCTs on its efficacy were first carried out over 100 years ago; iodised salt was on sale in the US in the 1920s, and in Europe before that; and the UN launched a campaign to eliminate iodine deficiency in 1990. I am no expert on this, any more than I am as regards iron as a treatment for anaemia. But if you would like to know more than BanerDu appear to, and learn something that is interesting in itself into the bargain, look for information on Helen Mackay, the first woman to receive the fellowship of the Royal College of Physicians of London, and her work in East London in the 1920s and 1930s. The take-home lesson here is that BanerDu, writing in a relaxed style than favours anecdote and chance observation, are working here pretty much entirely outside the spirit of the scientific protocols they claim to favour.
As it turns out, it is an essential part of their sedulous self-promotion that they never give credit to others when where it is due. So they discuss at length the finding that adding iron supplements to fish sauce in Indonesia enabled significant increases in the productivity and earnings of self-employed workers, but don't draw the reader's attention to a report of exactly the same finding from Basta et al (1979), discussed in the paper on which they draw (Thomas et al, 2004: 8). Similarly, as regards the debilitating effects of diarrhoea, hookworm and malaria that they address in the chapters on nutrition and health, both the conditions and the appropriate treatments are well understood, and have been for decades. They write, though, as if life, the universe and everything begins with the Poverty Action Lab. Regarding insecticide-impregnated bed nets as protection against malaria, they simply fail to mention the pioneering work supported by the WHO and carried out in the Gambia from the early 1990s - sophisticated scientific studies (carried out by my late friend Steve Bennett at the London Institute of Hygiene and Tropical Medicine among others), that incidentally explored in detail alternative methods of costing and distribution (communal as well as individual) to secure village-level and household-level engagement (Mills et al, 1995, Aikins et al, 1998, among many other references). And if you stay with bed nets as an example, you can see how the way they frame the issue limits from the start the value of the results of the experiment: BanerDu ask what difference it makes to take-up and use if the recipients have to pay for them, and find that in fact people value them just as much whether they pay for them or not. Good, but it matters where they are coming from. What they are in fact testing is whether people value things that are good for them if they get them for nothing, not what is the best way to slow or halt the spread of malaria. And this in turn is because they are working within the assumptions of an economic theory (much prized by advocates of user fees for everything) that says that what is free is not valued. Worse, the 'what is free is not valued' line is only cover for a prior ideological commitment to charging in principle for public goods, not because this will make people value them more but because it will force them to enter the market in order to earn the resources to pay for them. As to the efficacy of free distribution of bed nets, they run up immediately against the single variable problem that critics of RCT have identified, by leaving out the surrounding context (as noted above, not a mistake that was made in the Gambia). Does it matter who distributes the bed nets? What are the advantages and disadvantages of relying on communal institutions, or influential members of the community, versus medical practitioners or outside experts? What kind of information needs to accompany the nets? What needs to be done to make sure they are properly used? Should schools address the issue with the children who are users and beneficiaries of the nets? What other measures are needed to cope better with cases of malaria where they do occur? And as soon as you ask these questions, you can see that for a number of them, the answers will vary not only from country to country, but from locality to locality. Hence the consensus that the findings of RCTs do not entirely pin down causality, and cannot unproblematically be scaled up or assumed to hold in new locations.
The key to the book, in actual fact, is BanerDu's approach signature obsession with the productivity of the extremely poor. Taking as a point of reference an ideal situation in which the poor would conduct themselves in every instance in accordance with a logic of competitiveness and (human) capital accumulation, they focus on gaps between their actual behaviour and choices that would be conducive to their better integration into productive work and circuits of capitalist accumulation, and lifelong generation of surplus value for capital on the part of themselves and their children. Despite their strictures against the simple models of their fellow economists, they really want the poor to behave as 'rational' maximising individuals. So early on they ask why the poor wouldn't purchase cheap micro-nutrients, when doing so 'would probably make them, and almost certainly their children, significantly more successful in life?' (xx). They puzzle over the fact that when under-nourished workers experience an increase in income they don't devote the whole increase to buying extra calories, 'however monotonous and tasteless', when it would boost their potential energy output, productivity and earnings (25). Their take on the failure of parents in Kenya who refuse a few cents for deworming is that it is irrational because it deprives their children of 'hundreds of dollars of extra earning over their lifetime' (33). Micronutrients, likewise, are not only cheap, but 'can sometimes lead to a large increase in lifetime income' (34). So they take the propensity of the poor to seek occasionally to take some pleasure from life rather than to strive in every way to perfect themselves as machines at the service of capital as a challenge to their ingenuity. Why, they ask, would the poor rather have a television than a productivity-enhancing diet? You might wonder why they ask this at all, given the continual emphasis on how little it costs to buy essential micronutrients and extra calories. But they do. Their answer, a dazzling insight, is that 'things that make life less boring are a priority for the poor', a phenomenon perhaps explained by the 'basic human need for a pleasant life' (37). So: 'These "indulgences" are not the impulsive purchases of people who are not thinking hard about what they are doing. They are carefully thought out, and reflect strong compulsions, whether internally driven or externally imposed' (38). Lurking behind this is the old moralising middle class view of the poor as improvident (a view they later affect to question, p. 185), now dressed up in neuropsychological language. Pursuing this line of thought, BanerDu suggest that the poor 'often behave as if they think that any change that is significant enough to be worth sacrificing for will simply take too long'; may be because they are 'time-inconsistent', and don't 'properly' value the future against the present. So they favour 'nudging' approaches that exploit this - offering a bag of dhal, for example, to a mother who brings her child for immunisation, so the present incentive wins her over to the unduly discounted or unappreciated long term benefit (62-5). 'Nudges may be especially helpful when, for whatever reason, households are somewhat dubious about the benefits of what is being proposed to them' (67). Again, their explicit preference for a surreptitious nudge, rather than a sustained process of public education, betrays their condescending attitude to the poor.
This leads directly into a discussion of conditional cash transfers in relation to school attendance in Chapter Four (78-80), and the underlying objective soon becomes clear. Under the general headings of 'The Curse of Expectations' (parents expect too much), 'Why Schools Fail' (they focus too much on the better students) and 'Reengineering Education' (stick to the basics), BanerDu caution that parents who think that only higher level secondary schooling pays off and imagine their children going into government jobs are mistaken on both counts: in fact gains to future earnings are more or less proportional for each additional year of schooling, and government jobs are much harder to come by than parents think. But the 'good news, and it is very good news indeed, is that all the evidence we have strongly suggests that making sure that every child learns the basics well in school is not only possible, it is in fact fairly easy, as long as one focuses on doing exactly that, and nothing else' (97). Good news for capital, I suppose they mean, if the poor can be taught their place. They are quick to draw precisely that moral - that 'scaling down expectations, focusing on the core competencies, and using technology to complement, or if necessary substitute for, teachers' (100, emphasis mine) is the right path. And so to family planning, where again BanerDu are 100 per cent aligned with World Bank policy - focused on the need to reach women directly from early teenage years onwards, and ideally, where they are married, without their husbands' involvement. Of course, in itself this is a good thing. Consistently with what has gone before, though, their over-riding concern is with the strong link between delayed motherhood and participation in the formal labour market. So they report that 'women who had access to family planning as teenagers through [Colombia's Profamilia] programme had more schooling and were 7 percent more likely to work in the formal sector than those who did not' (110). They then profile some good work of their own in Kenya, and explore usefully and at some length the contrasting gendered interests within the 'black box' of the family. But this is in passing. The conclusion to the chapter, and to the first part of the book, captures well the way that everything is set squarely within the political economy of reform:
'The most effective population policy might therefore be to make it unnecessary to have so many children (in particular, so many male children). Effective social safety nets (such as health insurance or old age pensions) or even the kind of financial development that enables people to profitably save for retirement could lead to a substantial reduction in fertility and perhaps also less discrimination against girls. In the second part of the book, we turn to how this can be done' (128-9).
Before turning to the second part, then, we can summarise as follows the argument of the first: there are many ways in which the poor could boost their present and future productivity, if they only would. but they won't, so 'nudges' must be found, to increase their likelihood of doing so. They need to be taught basic skills that will equip them for the labour market, and their expectations need to be scaled down. Particular attention should be paid to young women, with a focus on measures that will delay their first pregnancy and allow them into the labour market, and all of this should be underpinned by public policies, of a kind to be explored.
BanerDu then introduce the second part of the book, on institutions, with the concept of risk, which was central to World Bank and other approaches to the reform of labour markets and social protection at the time, and remains so (Cammack, 2012): 'Risk is a central fact of life for the poor, who often run small businesses or farms or work as casual labourers, with no assurance of regular employment' (133). And like the World Bank, their aim is not to eliminate risk, but to drive or nudge the poor towards more productive risk frameworks. In successive chapters, BanerDu give accounts of insurance and finance markets, and the difficulties of providing for the very poor, and provide a measured assessment of microfinance (useful, but because of its strict conditions and conservative bias not particularly suited to promoting entrepreneurship). Their account of risk, though, is marked by unresolved contradictions, and a perverse approach to rationality which is governed in the end by a prior commitment to the abstract maximising individual. They set out initially with perfect clarity how poor operators of 'small businesses' and marginal farms live constantly on the edge of disaster, and how a single drought or economic shock can ruin them, giving at length the example of Ibu Tina, an Indonesian business woman driven into bankruptcy by one bad cheque. They persist in applying a medical/ psychological perspective to such cases of material disaster, even after recognising that Ibu Tina had been plunged into permanent poverty. As they comment: 'This process [of falling into permanent poverty] is often reinforced by a psychological process. Loss of hope and the sense that there is no easy way out can make it that much harder to have the self-control needed to try to climb back up the hill' (140). The sense that there is no easy way out [true of course, for all the extremely poor] is presented as a psychological condition that must be replaced by 'self-control'. Their diagnosis is that such characters 'did not seem to be in the mental shape needed to pick themselves up and start over'. Such bracing talk might well be appropriate for someone whose essay gets a 'B', or whose article gets rejected. But the case they go on to cite hilariously punctures their argument, although they barely acknowledge it:
'In Udaipur we met a man who said in response to a standard survey question that he had been so “worried, tense, or anxious” that it interfered with normal activities like sleeping, working, and eating for more than a month. We asked him why. He said that his camel had died, and he had been crying and tense ever since. Somewhat naïvely perhaps, we then went on to ask whether he had done something about his depression (like talk to a friend, a health-care practitioner, or a traditional healer). He seemed irked: “I have lost the camel. Of course I should be sad. There is nothing to be done”' (140).
He was polite, I think, in the circumstances. But despite their fleeting awareness that the man was entitled to despair at the sudden loss of his livelihood and the prospect of indigence, they still insist on framing the issue in psychological terms. The paragraph begins with the statement cited above that impoverishment is often reinforced by a psychological process, and the paragraph immediately following begins, 'There may be other psychological forces at work as well'. They go on to sketch out perfectly clearly the propensity of poor people to hedge by maintaining a number of occupations and sources of income and the logic behind it, but even then cannot clear from their minds the image of the maximising individual who trades, Ricardo-style, on comparative advantage. Having multiple occupations, they conclude, is inefficient, as it obliges people to 'pass up the gains from specializing in what they are really good at' (143). Wise advice, no doubt, for someone who is 'good at' intermittently available casual labour.
In the chapters that follow, on saving and jobs, the ideological foundations and underlying contradictions of their position emerge ever more sharply. On saving, they note the relative paucity of savings options for the poor, but the focus soon shifts to the 'nascent capitalist inside every poor man and woman' (185), and how this acquisitive individual can be brought to life. Chennai fruit vendors - petty 'entrepreneurs' on the breadline - borrow money each morning to buy their stock. If only they would buy two fewer cups of tea per day for just three days, they calculate, and thereby make possible a permanent reduction in the loan they need each day, they would be debt-free in ninety days, and save 40 rupees interest per day. 'The point,' they conclude in one of their most bizarre asides 'is that these vendors are sitting under what appears to be as close to a money tree as we are likely to find anywhere (190-91). This leads them straight back to the psychology and neuroscience of self-control, and the need to devise ways to protect the poor against themselves by tying their hands over current expenditure in various ways, and we find ourselves, despite their half-hearted attempts to take distance from it, on ground that was familiar in Marx's day:
'What is the quality in which the improvident classes of this country are so deficient as self-denial, — the ability to sacrifice a small present gratification for a future good? Those classes who work the hardest might naturally be expected to value the most the money which they earn. Yet the readiness with which so many are accustomed to eat up and drink up their earnings as they go, renders them to a great extent absolutely helpless and dependent upon the frugal' (Samuel Smiles, Self-Help, 1859, Chapter 9).
It is no surprise that this classic Victorian text is made available online by the Liberty Fund. Smiles expressed his view of capitalism and of the feckless poor in what were by then conventional religious terms: 'That there should be a class of men who live by their daily labor in every state is the ordinance of God, and doubtless is a wise and righteous one; but that this class should be otherwise than frugal, contented, intelligent, and happy, is not the design of Providence, but springs solely from the weakness, self-indulgence, and perverseness of man himself'. Like BanerDu, Smiles was fond of sturdy proletarian examplars - Thomas Wright, for example, who worked twelve hours a day for six days a week in a foundry, provided for his family, and devoted his Sundays to working with convicted criminals, rescuing 'not fewer than three hundred felons from continuance in a life of villany!' (ibid). But where Smiles thought 'mechanical aids, such as pledges' inferior to the adoption of 'a high standard of thinking and acting', BanerDu have made a profession of dreaming them up. When their schemes don't work, they reach for a normative interpretation even when the material facts are staring them in the face. So, reporting on an experiment with fruit vendors in India and the Philippines, which paid off their debts in full:
For a while, many of the vendors managed to stay debt-free: After ten weeks, 40 percent were still debt-free in the Philippines. So these fruit vendors seem to have enough patience to stay out of debt for a while. On the other hand, almost all of them eventually fell back into debt. It was usually a shock (an illness, and emergency need) that pushed them back into debt, and once that happened, they did not manage to pay the debt back on their own. This asymmetry between managing to stay free of debt and not managing to get out of debt shows the role of discouragement in making it harder to impose self-discipline' (202).
The subterfuge (or, who knows, unconscious bias) is transparent. The material fact that for micro-entrepreneurs with no margin of financial safety a sudden shock throws them back into debt is converted into a moral drama where after a period patience gives way to discouragement. The following chapter (9 - Reluctant Entrepreneurs) opens with a reductio ad absurdum which I leave to you to explore, but moves by degrees to the summary statement that: 'First, while many of the poor operate businesses, they mainly operate tiny businesses. And second, these tiny businesses are, for the most part, making very little money'; 'the vast majority of the businesses run by the poor never grow to the point where they start having any employees or much in the way of assets' (213). It has taken them a long time to get to this essential point, and it leads directly to the conclusion that 'If the businesses run by the poor are generally unprofitable, this may well explain why giving them a loan to start a new business does not lead to a drastic improvement in their welfare' (214) - a statement that in turn calls into question a very large part of what has gone before. The message is reinforced a few pages later:
'This is the paradox of the poor and their businesses: They are energetic and resourceful and manage to make a lot out of a very little. But most of this energy is spent on businesses that are too small and utterly undifferentiated from the many others around them. As a result, their operators have no chance to earn a reasonable living' (218).
It takes BanerDu this long, then, to reach the conclusion that businesses only thrive if they reach the point - inaccessible to the vast majority of the extremely poor - where they can invest heavily in advanced and expensive production technologies, and build a sizeable workforce. For 'businesses' that are loss-making if labour time is costed at the level of the minimum wage, and face competition from a myriad similar local competitors, this is not possible. Hence the tendency to invest in two or three micro-enterprises, to squeeze scraps of income out of 'free' time. So they conclude:
'Taken together, this evidence makes us seriously doubt the idea that the average small business owner is a natural "entrepreneur," in the way we generally understand the term, meaning someone whose business has the potential to grow and who is able to take risks, work hard, and keep trying to make it happen even in the face of multiple hardships' (225).
In itself, this is a perfectly sound conclusion. But at the same time, it is one that owes nothing to any randomised controlled trial, renders most of the book to this point superfluous, and constitutes a 'radical rethinking' only in relation to anyone gripped by the illusion that extreme poverty mysteriously harbours the key ingredients for business success. From here it is only a short step to the thinking prevalent in the World Bank at precisely this time, and reflected in the notion of 'good jobs'. First, BanerDu summarise that: 'Perhaps the many businesses of the poor are less a testimony to their entrepreneurial spirit than a symptom of the dramatic failure of the economies in which they live to provide them with something better' (226), then in a section entitled simply 'Good Jobs' (226-34) that runs to the end of the chapter, that make precisely the case advanced in the 2013 World Development Report: Jobs, in preparation at the time and published in 2012. And indeed, Jobs cites Poor Economics on four occasions, each time to make the point, one way or another, that micro-enterprises and household businesses 'are a means of survival for the poor and a way of diversifying out of farming activities. On average, their owners do not earn much' (p. 12; cf. 56, 110, 114). BanerDu's call for 'good jobs' - ideally stable jobs in productive enterprises linked through exports to the world market - coincides exactly with the message of the Report, and advocates such familiar policies as effective social safety nets, labour market reforms, infrastructural investment, and loans to medium-sized businesses (232-3). After all this, though, the final substantive chapter, 'Policies, Politics' says nothing about this agenda, but instead defends RCTs against the criticism that they lack ambition, runs through a range of secondary literature, and settles on reform at the micro-institutional level as the way forward. What is striking, as before, is the narrow range of reference, centred as always on BanerDu and associates, and crying out for some acknowledgement at least of alternative approaches such as that reflected in Judith Tendler's Good Government in the Tropics (Johns Hopkins, 1997). As always, BanerDu trade too much on a simplistic contrast between macro- and micro-levels of reform, making out that there is nothing in between.
Not surprisingly, their conclusions are cursory, and anodyne in the extreme: 'First, the poor often lack critical pieces of information and believe things that are not true' (267); 'Second, the poor bear responsibility for too many aspects of their lives' (268); 'Third, there are good reasons that some markets are missing for the poor, or that the poor face unfavorable prices in them' (269); 'Fourth, poor countries are not doomed to failure because they are poor, or because they have an unfortunate history' (270); 'Finally, expectations about what people are able or unable to do all too often end up turning into self-fulfilling prophecies' (271).
In short, a truly awful book. It remains to seen if their next (Good Economics for Hard Times) is any better.
References
Aikins, Moses Kweku et al, 1998. 'The Gambian National Impregnated Bednet Programme: Costs, consequences and net cost-effectiveness', Social Science & Medicine, 46, 2, 181-91.
Banerjee, Abhijit, and Esther Duflo, 2009. 'The Experimental Approach to Development Economics,' Annual Review of Economics, 1, 151-178.
Barrett, Christopher B, and Michael R. Carter, 2010. '“The Power and Pitfalls of Experiments in Development Economics: Some Non-Random Reflections,' Applied Economic Perspectives and Policy, 32, 4.
Basta, S, K. Soekirman and N Scrimshaw, 1979. 'Iron deficiency anemia and productivity of adult males in Indonesia', American Journal of Clinical Nutrition, 32, 916-25.
Cammack, Paul, 2012. ‘Risk, Social Protection and the World Market’, Journal of Contemporary Asia, 42, 3, 359-377.
Cartwright, Nancy, 2010. 'What Are Randomised Controlled Trials Good For?', Philosophical Studies, 147, 1, 59–70.
Deaton, Angus and Nancy Cartwright, 2018. 'Understanding and misunderstanding randomized controlled trials', Social Science and Medicine, 210, 2–21.
Heckman, James and Jeffrey Smith, 1995. 'Assessing the Case for Social Experiments,' Journal of Economic Perspectives, 9, 2, pp. 85-110.
Kremer, Michael, Gautam Rao and Frank Schilbach, 'Behavioral Development Economics', in B. Douglas Bernheim, Stefano DellaVigna, David Laibson, eds, Handbook of Behavioral Economics: Foundations and Applications, 2, 345-458.
Mills, Anne, et al. 1995. 'Financing mechanisms for village activities in The Gambia and their implications for financing insecticide for bed net impregnation', Journal of Tropical Medicine and Hygiene, 97, 6, 325-32.
Ravallion, Martin, 2009. 'Should the Randomistas Rule?', Economists’ Voice, 6, 2, 1-5.
Ravallion, Martin, 2012. 'Fighting Poverty One Experiment at a Time: A Review of Abhijit Banerjee and Esther Duflo's Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty', Journal of Economic Literature, 50, 1, 103-114.
Ravallion, Martin, 2018. 'Should the Randomistas (Continue To) Rule?', Working Paper 492, Center for Global Development, New York (January 2019 update).
Ravallion 2020. 'Highly Prized Experiments', World Development, 127, 2020.
Reddy, Sanjay G. 2012. 'Randomise This! On Poor Economics', Review of Agrarian Studies, 2, 2.
Rodrik, Dani, 2009. 'The New Development Economics: We Shall Experiment, but How Shall We Learn?, in Jessica Cohen and William Easterly, eds, What Works in Development? Thinking Big and Thinking Small, Brookings Institution Press, Washington DC.
Thomas, Duncan, Elizabeth Frankenberg, Jed Friedman, et al., 2004. 'Causal Effect of Health on Labor Market Outcomes: Evidence from a Random Assignment Iron Supplementation Intervention', mimeo (available at https://escholarship.org/uc/item/1h66k92r).
A distinction also needs to be made between randomised controlled trials (RCTs) on the one hand, and Poor Economics on the other. There is a substantial critical literature on RCTs - 'lab' or field 'experiments' modelled on the method by which new drugs are tested, through which the prize recipients have made their reputations - some of it predating the publication of Poor Economics (notably, Heckman and Smith, 1995; Ravallion, 2009; Rodrik, 2009; Barrett and Carter, 2010; Cartwright, 2010), some following its publication (notably Ravallion, 2012, Reddy, 2012, Deaton and Cartwright, 2018, Ravallion 2018), and some following the award of the prize (notably, the dossier of short responses in World Development, 127, 2020, among which the persistent Ravallion, 2020). Together and independently, the prize recipients have also contributed to the debate (e.g. Duflo and Kremer, 2008; Banerjee and Duflo, 2009; Banerjee, Duflo and Kremer, 2019). The consensus of this literature (not shared by the most enthusiastic) is that RCTs represent a useful addition to techniques of evaluation, and that they work better with specific and delimited issues than with broad policy initiatives, but that they do not either escape all the problems that other methods face, or lend themselves unproblematically to scaling up, transfer to different locations, or estimates of long-term consequences. They do not resolve definitively issues of causality or efficacy, and there is a general concern that they are more beset by ethical issues than their counterparts in the medical world, but far less advanced in the development of protocols to address them.
All this, however, is largely tangential to Poor Economics, which is not only about RCTs, and in which none of the critiques of RCTs available when it was written is mentioned, let alone addressed. It is not a contribution to the debate, or even a scholarly exposition of BanerDu's approach. Rather, it is a piece of shameless self-promotion which gives precious little credit to the work of others, and which, despite the claim of 'radical rethinking', follows an entirely orthodox line of analysis derived from the World Bank of the 1990s. Far from being scientific in any sense, it is ideological to the core. Perhaps out of a sense of professional decorum, even the reviews of the book itself cited above (Ravallion, 2012; Reddy, 2012) don't tell you just what an appalling piece of hagiography it is.
From the start, its true heroes are not RCTs, or the extreme poor, but Banerjee and Duflo (hereafter BanerDu) themselves. Poor Economics invokes Esther and Abhijit, aged six, on the very first page (along with Mother Teresa!), hinting at a lifelong and indeed a saintly vocation. It then launches into a diatribe that accuses the development economics and policy professions at large of reducing the poor to cartoon characters, with no names given, or examples provided. 'Because the poor possess very little,' they say, 'it is assumed that there is nothing interesting about their economic existence' (viii). By whom? And what could this possibly mean? This is writing of the most light-minded kind, without any discernible thought process attached, and it sets the tone. BanerDu claim, in contrast to their peers, to have always taken 'the time to really understand [the] lives [of the poor], in all their complexity and richness'; they have been to the 'back alleys and villages where the poor live', and unlike those unspecified other '(often Western or Western-trained) professional development economists and policy makers', with the 'simple models' they have traditionally used to think about the lives of the poor, they have struggled through initial confusion to 'knit together a coherent story of how poor people live their lives' (viii-ix). In fact they use a simple model themselves (the S-curve, which addresses the possibility of poverty traps), and their ethnography, such as it is, is shallow and anecdotal; and Banerjee (Ph.D. Harvard, 1988, subsequently at Princeton, Harvard and MIT) and Duflo (Ph.D. MIT, 1999, and there ever since) are quintessentially mainstream development economists themselves. 'What is striking,' they say all the same of people on less than a dollar a day, in another witless pronouncement, 'is that even people who are that poor are just like the rest of us in almost every way' (ix), so managing not only to patronise them but also to dismiss entirely the significance of class difference. You know right away that you are in for an analysis that will be entirely uncritical of a system in which a small minority own the means of production, and the great majority are dependent on their capacity to work to survive. It quickly becomes clear, too, that they do not simply want to understand the poor. They want to change them. In a more accurate description of their purpose they say that this is 'a book about the kinds of theories that help us make sense of both what the poor are able to achieve, and where and for what reason they need a push' (x). And in the closing lines of the foreword, they give an early indication of the kind of push they need (one that will be familiar to anyone acquainted with World Bank policy):
'The book ... tells a lot about where hope lies: why token subsidies might have more than token effects; how to market better insurance; why less may be more in education; why good jobs matter for growth' (xi).
Everything, it turns out, is about making the global labour market work better.
The first step BanerDu take in embarking on their journey is to sideline what they call 'the "big questions": What is the ultimate cause of poverty? How much faith should we place in free markets? Is democracy good for the poor? Does foreign aid have a role to play? And so on' (3). They discuss only the last (and smallest) of these, and then only to say that there is no simple answer. But don't be deceived. This position on 'big questions' is by no means as naive as it looks, as they only affect agnosticism. Taking the first two in reverse order, they place their faith uncritically, not in entirely free markets, but in the private enterprise system, suitably regulated. Specifically, they endorse the rule of capital, and seek to turn the extremely poor into the productive poor, in part and with increasing reservations through petty entrepreneurship, but primarily, though late on and without extended discussion, through the provision of 'good jobs'. By affecting to eschew 'big questions' they bracket this preference off from debate. This leaves them by default with a clear and clearly inadequate position on the cause of poverty - it arises from mistaken policies on the part of well-intentioned governments, and poor choices by the poor, who act rationally within the limits of their own understanding (though with exceptions common to all of us, such as present bias and procrastination, which can be bracketed off in turn as 'human nature'), but lack the knowledge and resources to act in their own best long term interest as BanerDu perceive it.
All this means that there is very little that is new in their policy preferences. They are entirely in tune with such former Nobel prizewinners as Amartya Sen (1998), Joseph Stiglitz (2001) and Richard 'Nudge' Thaler (2017), and hand-in-glove with the leading institutions that manage the global capitalist economy. Their agenda echoes that of the World Bank point for point, and is equally oriented towards reforms that change behaviour by altering incentive structures. The very first World Development Report, published coincidentally in 1978, when Duflo was six years old and fixated on Mother Teresa, called for action on 'the structure of economic incentives, the allocation of investments, and the creation of special institutions and programs to increase the productivity of the poor and their opportunities for employment’ (World Bank, 1978: 65). In 1980 the Bank expanded in detail on nutrition, health, education and population control as key components of a strategy to increase the productivity of the poor (World Bank, 1980: 46-70). This is the water in which BanerDu have been swimming since they first dived into development studies, and always with the current. So much so, in fact, that the first four chapters of Poor Economics cover in turn nutrition, health, education, and population control - a full house, albeit in a four-card hand. Nor are the specific problems and potential solutions they take up new. So, for example, the first focuses on the efficacy of readily available and inexpensive nutritional supplements, such as iron and iodine. And although they repeatedly claim that (again unspecified) development economists and agencies continue to neglect such micro-nutrients, arguing instead that the problem is a deficiency of protein and calories, these precise micro-nutritional deficiencies - iron and iodine - were highlighted in the 1980 World Development Report itself (p. 63), and have been the focus of attention ever since. BanerDu must be well aware of this, just as they show themselves to be aware that periodic food 'shortages' do occur, arising from combinations of crop failure and maldistribution, and do lead to serious famines from time to time. Mystifyingly, they remark that: 'The importance of micronutrients was not fully understood, even by scientists, until relatively recently' (34). I'm not so sure about that. The health-related properties of iodised salt were identified in the 1830s; medical RCTs on its efficacy were first carried out over 100 years ago; iodised salt was on sale in the US in the 1920s, and in Europe before that; and the UN launched a campaign to eliminate iodine deficiency in 1990. I am no expert on this, any more than I am as regards iron as a treatment for anaemia. But if you would like to know more than BanerDu appear to, and learn something that is interesting in itself into the bargain, look for information on Helen Mackay, the first woman to receive the fellowship of the Royal College of Physicians of London, and her work in East London in the 1920s and 1930s. The take-home lesson here is that BanerDu, writing in a relaxed style than favours anecdote and chance observation, are working here pretty much entirely outside the spirit of the scientific protocols they claim to favour.
As it turns out, it is an essential part of their sedulous self-promotion that they never give credit to others when where it is due. So they discuss at length the finding that adding iron supplements to fish sauce in Indonesia enabled significant increases in the productivity and earnings of self-employed workers, but don't draw the reader's attention to a report of exactly the same finding from Basta et al (1979), discussed in the paper on which they draw (Thomas et al, 2004: 8). Similarly, as regards the debilitating effects of diarrhoea, hookworm and malaria that they address in the chapters on nutrition and health, both the conditions and the appropriate treatments are well understood, and have been for decades. They write, though, as if life, the universe and everything begins with the Poverty Action Lab. Regarding insecticide-impregnated bed nets as protection against malaria, they simply fail to mention the pioneering work supported by the WHO and carried out in the Gambia from the early 1990s - sophisticated scientific studies (carried out by my late friend Steve Bennett at the London Institute of Hygiene and Tropical Medicine among others), that incidentally explored in detail alternative methods of costing and distribution (communal as well as individual) to secure village-level and household-level engagement (Mills et al, 1995, Aikins et al, 1998, among many other references). And if you stay with bed nets as an example, you can see how the way they frame the issue limits from the start the value of the results of the experiment: BanerDu ask what difference it makes to take-up and use if the recipients have to pay for them, and find that in fact people value them just as much whether they pay for them or not. Good, but it matters where they are coming from. What they are in fact testing is whether people value things that are good for them if they get them for nothing, not what is the best way to slow or halt the spread of malaria. And this in turn is because they are working within the assumptions of an economic theory (much prized by advocates of user fees for everything) that says that what is free is not valued. Worse, the 'what is free is not valued' line is only cover for a prior ideological commitment to charging in principle for public goods, not because this will make people value them more but because it will force them to enter the market in order to earn the resources to pay for them. As to the efficacy of free distribution of bed nets, they run up immediately against the single variable problem that critics of RCT have identified, by leaving out the surrounding context (as noted above, not a mistake that was made in the Gambia). Does it matter who distributes the bed nets? What are the advantages and disadvantages of relying on communal institutions, or influential members of the community, versus medical practitioners or outside experts? What kind of information needs to accompany the nets? What needs to be done to make sure they are properly used? Should schools address the issue with the children who are users and beneficiaries of the nets? What other measures are needed to cope better with cases of malaria where they do occur? And as soon as you ask these questions, you can see that for a number of them, the answers will vary not only from country to country, but from locality to locality. Hence the consensus that the findings of RCTs do not entirely pin down causality, and cannot unproblematically be scaled up or assumed to hold in new locations.
The key to the book, in actual fact, is BanerDu's approach signature obsession with the productivity of the extremely poor. Taking as a point of reference an ideal situation in which the poor would conduct themselves in every instance in accordance with a logic of competitiveness and (human) capital accumulation, they focus on gaps between their actual behaviour and choices that would be conducive to their better integration into productive work and circuits of capitalist accumulation, and lifelong generation of surplus value for capital on the part of themselves and their children. Despite their strictures against the simple models of their fellow economists, they really want the poor to behave as 'rational' maximising individuals. So early on they ask why the poor wouldn't purchase cheap micro-nutrients, when doing so 'would probably make them, and almost certainly their children, significantly more successful in life?' (xx). They puzzle over the fact that when under-nourished workers experience an increase in income they don't devote the whole increase to buying extra calories, 'however monotonous and tasteless', when it would boost their potential energy output, productivity and earnings (25). Their take on the failure of parents in Kenya who refuse a few cents for deworming is that it is irrational because it deprives their children of 'hundreds of dollars of extra earning over their lifetime' (33). Micronutrients, likewise, are not only cheap, but 'can sometimes lead to a large increase in lifetime income' (34). So they take the propensity of the poor to seek occasionally to take some pleasure from life rather than to strive in every way to perfect themselves as machines at the service of capital as a challenge to their ingenuity. Why, they ask, would the poor rather have a television than a productivity-enhancing diet? You might wonder why they ask this at all, given the continual emphasis on how little it costs to buy essential micronutrients and extra calories. But they do. Their answer, a dazzling insight, is that 'things that make life less boring are a priority for the poor', a phenomenon perhaps explained by the 'basic human need for a pleasant life' (37). So: 'These "indulgences" are not the impulsive purchases of people who are not thinking hard about what they are doing. They are carefully thought out, and reflect strong compulsions, whether internally driven or externally imposed' (38). Lurking behind this is the old moralising middle class view of the poor as improvident (a view they later affect to question, p. 185), now dressed up in neuropsychological language. Pursuing this line of thought, BanerDu suggest that the poor 'often behave as if they think that any change that is significant enough to be worth sacrificing for will simply take too long'; may be because they are 'time-inconsistent', and don't 'properly' value the future against the present. So they favour 'nudging' approaches that exploit this - offering a bag of dhal, for example, to a mother who brings her child for immunisation, so the present incentive wins her over to the unduly discounted or unappreciated long term benefit (62-5). 'Nudges may be especially helpful when, for whatever reason, households are somewhat dubious about the benefits of what is being proposed to them' (67). Again, their explicit preference for a surreptitious nudge, rather than a sustained process of public education, betrays their condescending attitude to the poor.
This leads directly into a discussion of conditional cash transfers in relation to school attendance in Chapter Four (78-80), and the underlying objective soon becomes clear. Under the general headings of 'The Curse of Expectations' (parents expect too much), 'Why Schools Fail' (they focus too much on the better students) and 'Reengineering Education' (stick to the basics), BanerDu caution that parents who think that only higher level secondary schooling pays off and imagine their children going into government jobs are mistaken on both counts: in fact gains to future earnings are more or less proportional for each additional year of schooling, and government jobs are much harder to come by than parents think. But the 'good news, and it is very good news indeed, is that all the evidence we have strongly suggests that making sure that every child learns the basics well in school is not only possible, it is in fact fairly easy, as long as one focuses on doing exactly that, and nothing else' (97). Good news for capital, I suppose they mean, if the poor can be taught their place. They are quick to draw precisely that moral - that 'scaling down expectations, focusing on the core competencies, and using technology to complement, or if necessary substitute for, teachers' (100, emphasis mine) is the right path. And so to family planning, where again BanerDu are 100 per cent aligned with World Bank policy - focused on the need to reach women directly from early teenage years onwards, and ideally, where they are married, without their husbands' involvement. Of course, in itself this is a good thing. Consistently with what has gone before, though, their over-riding concern is with the strong link between delayed motherhood and participation in the formal labour market. So they report that 'women who had access to family planning as teenagers through [Colombia's Profamilia] programme had more schooling and were 7 percent more likely to work in the formal sector than those who did not' (110). They then profile some good work of their own in Kenya, and explore usefully and at some length the contrasting gendered interests within the 'black box' of the family. But this is in passing. The conclusion to the chapter, and to the first part of the book, captures well the way that everything is set squarely within the political economy of reform:
'The most effective population policy might therefore be to make it unnecessary to have so many children (in particular, so many male children). Effective social safety nets (such as health insurance or old age pensions) or even the kind of financial development that enables people to profitably save for retirement could lead to a substantial reduction in fertility and perhaps also less discrimination against girls. In the second part of the book, we turn to how this can be done' (128-9).
Before turning to the second part, then, we can summarise as follows the argument of the first: there are many ways in which the poor could boost their present and future productivity, if they only would. but they won't, so 'nudges' must be found, to increase their likelihood of doing so. They need to be taught basic skills that will equip them for the labour market, and their expectations need to be scaled down. Particular attention should be paid to young women, with a focus on measures that will delay their first pregnancy and allow them into the labour market, and all of this should be underpinned by public policies, of a kind to be explored.
BanerDu then introduce the second part of the book, on institutions, with the concept of risk, which was central to World Bank and other approaches to the reform of labour markets and social protection at the time, and remains so (Cammack, 2012): 'Risk is a central fact of life for the poor, who often run small businesses or farms or work as casual labourers, with no assurance of regular employment' (133). And like the World Bank, their aim is not to eliminate risk, but to drive or nudge the poor towards more productive risk frameworks. In successive chapters, BanerDu give accounts of insurance and finance markets, and the difficulties of providing for the very poor, and provide a measured assessment of microfinance (useful, but because of its strict conditions and conservative bias not particularly suited to promoting entrepreneurship). Their account of risk, though, is marked by unresolved contradictions, and a perverse approach to rationality which is governed in the end by a prior commitment to the abstract maximising individual. They set out initially with perfect clarity how poor operators of 'small businesses' and marginal farms live constantly on the edge of disaster, and how a single drought or economic shock can ruin them, giving at length the example of Ibu Tina, an Indonesian business woman driven into bankruptcy by one bad cheque. They persist in applying a medical/ psychological perspective to such cases of material disaster, even after recognising that Ibu Tina had been plunged into permanent poverty. As they comment: 'This process [of falling into permanent poverty] is often reinforced by a psychological process. Loss of hope and the sense that there is no easy way out can make it that much harder to have the self-control needed to try to climb back up the hill' (140). The sense that there is no easy way out [true of course, for all the extremely poor] is presented as a psychological condition that must be replaced by 'self-control'. Their diagnosis is that such characters 'did not seem to be in the mental shape needed to pick themselves up and start over'. Such bracing talk might well be appropriate for someone whose essay gets a 'B', or whose article gets rejected. But the case they go on to cite hilariously punctures their argument, although they barely acknowledge it:
'In Udaipur we met a man who said in response to a standard survey question that he had been so “worried, tense, or anxious” that it interfered with normal activities like sleeping, working, and eating for more than a month. We asked him why. He said that his camel had died, and he had been crying and tense ever since. Somewhat naïvely perhaps, we then went on to ask whether he had done something about his depression (like talk to a friend, a health-care practitioner, or a traditional healer). He seemed irked: “I have lost the camel. Of course I should be sad. There is nothing to be done”' (140).
He was polite, I think, in the circumstances. But despite their fleeting awareness that the man was entitled to despair at the sudden loss of his livelihood and the prospect of indigence, they still insist on framing the issue in psychological terms. The paragraph begins with the statement cited above that impoverishment is often reinforced by a psychological process, and the paragraph immediately following begins, 'There may be other psychological forces at work as well'. They go on to sketch out perfectly clearly the propensity of poor people to hedge by maintaining a number of occupations and sources of income and the logic behind it, but even then cannot clear from their minds the image of the maximising individual who trades, Ricardo-style, on comparative advantage. Having multiple occupations, they conclude, is inefficient, as it obliges people to 'pass up the gains from specializing in what they are really good at' (143). Wise advice, no doubt, for someone who is 'good at' intermittently available casual labour.
In the chapters that follow, on saving and jobs, the ideological foundations and underlying contradictions of their position emerge ever more sharply. On saving, they note the relative paucity of savings options for the poor, but the focus soon shifts to the 'nascent capitalist inside every poor man and woman' (185), and how this acquisitive individual can be brought to life. Chennai fruit vendors - petty 'entrepreneurs' on the breadline - borrow money each morning to buy their stock. If only they would buy two fewer cups of tea per day for just three days, they calculate, and thereby make possible a permanent reduction in the loan they need each day, they would be debt-free in ninety days, and save 40 rupees interest per day. 'The point,' they conclude in one of their most bizarre asides 'is that these vendors are sitting under what appears to be as close to a money tree as we are likely to find anywhere (190-91). This leads them straight back to the psychology and neuroscience of self-control, and the need to devise ways to protect the poor against themselves by tying their hands over current expenditure in various ways, and we find ourselves, despite their half-hearted attempts to take distance from it, on ground that was familiar in Marx's day:
'What is the quality in which the improvident classes of this country are so deficient as self-denial, — the ability to sacrifice a small present gratification for a future good? Those classes who work the hardest might naturally be expected to value the most the money which they earn. Yet the readiness with which so many are accustomed to eat up and drink up their earnings as they go, renders them to a great extent absolutely helpless and dependent upon the frugal' (Samuel Smiles, Self-Help, 1859, Chapter 9).
It is no surprise that this classic Victorian text is made available online by the Liberty Fund. Smiles expressed his view of capitalism and of the feckless poor in what were by then conventional religious terms: 'That there should be a class of men who live by their daily labor in every state is the ordinance of God, and doubtless is a wise and righteous one; but that this class should be otherwise than frugal, contented, intelligent, and happy, is not the design of Providence, but springs solely from the weakness, self-indulgence, and perverseness of man himself'. Like BanerDu, Smiles was fond of sturdy proletarian examplars - Thomas Wright, for example, who worked twelve hours a day for six days a week in a foundry, provided for his family, and devoted his Sundays to working with convicted criminals, rescuing 'not fewer than three hundred felons from continuance in a life of villany!' (ibid). But where Smiles thought 'mechanical aids, such as pledges' inferior to the adoption of 'a high standard of thinking and acting', BanerDu have made a profession of dreaming them up. When their schemes don't work, they reach for a normative interpretation even when the material facts are staring them in the face. So, reporting on an experiment with fruit vendors in India and the Philippines, which paid off their debts in full:
For a while, many of the vendors managed to stay debt-free: After ten weeks, 40 percent were still debt-free in the Philippines. So these fruit vendors seem to have enough patience to stay out of debt for a while. On the other hand, almost all of them eventually fell back into debt. It was usually a shock (an illness, and emergency need) that pushed them back into debt, and once that happened, they did not manage to pay the debt back on their own. This asymmetry between managing to stay free of debt and not managing to get out of debt shows the role of discouragement in making it harder to impose self-discipline' (202).
The subterfuge (or, who knows, unconscious bias) is transparent. The material fact that for micro-entrepreneurs with no margin of financial safety a sudden shock throws them back into debt is converted into a moral drama where after a period patience gives way to discouragement. The following chapter (9 - Reluctant Entrepreneurs) opens with a reductio ad absurdum which I leave to you to explore, but moves by degrees to the summary statement that: 'First, while many of the poor operate businesses, they mainly operate tiny businesses. And second, these tiny businesses are, for the most part, making very little money'; 'the vast majority of the businesses run by the poor never grow to the point where they start having any employees or much in the way of assets' (213). It has taken them a long time to get to this essential point, and it leads directly to the conclusion that 'If the businesses run by the poor are generally unprofitable, this may well explain why giving them a loan to start a new business does not lead to a drastic improvement in their welfare' (214) - a statement that in turn calls into question a very large part of what has gone before. The message is reinforced a few pages later:
'This is the paradox of the poor and their businesses: They are energetic and resourceful and manage to make a lot out of a very little. But most of this energy is spent on businesses that are too small and utterly undifferentiated from the many others around them. As a result, their operators have no chance to earn a reasonable living' (218).
It takes BanerDu this long, then, to reach the conclusion that businesses only thrive if they reach the point - inaccessible to the vast majority of the extremely poor - where they can invest heavily in advanced and expensive production technologies, and build a sizeable workforce. For 'businesses' that are loss-making if labour time is costed at the level of the minimum wage, and face competition from a myriad similar local competitors, this is not possible. Hence the tendency to invest in two or three micro-enterprises, to squeeze scraps of income out of 'free' time. So they conclude:
'Taken together, this evidence makes us seriously doubt the idea that the average small business owner is a natural "entrepreneur," in the way we generally understand the term, meaning someone whose business has the potential to grow and who is able to take risks, work hard, and keep trying to make it happen even in the face of multiple hardships' (225).
In itself, this is a perfectly sound conclusion. But at the same time, it is one that owes nothing to any randomised controlled trial, renders most of the book to this point superfluous, and constitutes a 'radical rethinking' only in relation to anyone gripped by the illusion that extreme poverty mysteriously harbours the key ingredients for business success. From here it is only a short step to the thinking prevalent in the World Bank at precisely this time, and reflected in the notion of 'good jobs'. First, BanerDu summarise that: 'Perhaps the many businesses of the poor are less a testimony to their entrepreneurial spirit than a symptom of the dramatic failure of the economies in which they live to provide them with something better' (226), then in a section entitled simply 'Good Jobs' (226-34) that runs to the end of the chapter, that make precisely the case advanced in the 2013 World Development Report: Jobs, in preparation at the time and published in 2012. And indeed, Jobs cites Poor Economics on four occasions, each time to make the point, one way or another, that micro-enterprises and household businesses 'are a means of survival for the poor and a way of diversifying out of farming activities. On average, their owners do not earn much' (p. 12; cf. 56, 110, 114). BanerDu's call for 'good jobs' - ideally stable jobs in productive enterprises linked through exports to the world market - coincides exactly with the message of the Report, and advocates such familiar policies as effective social safety nets, labour market reforms, infrastructural investment, and loans to medium-sized businesses (232-3). After all this, though, the final substantive chapter, 'Policies, Politics' says nothing about this agenda, but instead defends RCTs against the criticism that they lack ambition, runs through a range of secondary literature, and settles on reform at the micro-institutional level as the way forward. What is striking, as before, is the narrow range of reference, centred as always on BanerDu and associates, and crying out for some acknowledgement at least of alternative approaches such as that reflected in Judith Tendler's Good Government in the Tropics (Johns Hopkins, 1997). As always, BanerDu trade too much on a simplistic contrast between macro- and micro-levels of reform, making out that there is nothing in between.
Not surprisingly, their conclusions are cursory, and anodyne in the extreme: 'First, the poor often lack critical pieces of information and believe things that are not true' (267); 'Second, the poor bear responsibility for too many aspects of their lives' (268); 'Third, there are good reasons that some markets are missing for the poor, or that the poor face unfavorable prices in them' (269); 'Fourth, poor countries are not doomed to failure because they are poor, or because they have an unfortunate history' (270); 'Finally, expectations about what people are able or unable to do all too often end up turning into self-fulfilling prophecies' (271).
In short, a truly awful book. It remains to seen if their next (Good Economics for Hard Times) is any better.
References
Aikins, Moses Kweku et al, 1998. 'The Gambian National Impregnated Bednet Programme: Costs, consequences and net cost-effectiveness', Social Science & Medicine, 46, 2, 181-91.
Banerjee, Abhijit, and Esther Duflo, 2009. 'The Experimental Approach to Development Economics,' Annual Review of Economics, 1, 151-178.
Barrett, Christopher B, and Michael R. Carter, 2010. '“The Power and Pitfalls of Experiments in Development Economics: Some Non-Random Reflections,' Applied Economic Perspectives and Policy, 32, 4.
Basta, S, K. Soekirman and N Scrimshaw, 1979. 'Iron deficiency anemia and productivity of adult males in Indonesia', American Journal of Clinical Nutrition, 32, 916-25.
Cammack, Paul, 2012. ‘Risk, Social Protection and the World Market’, Journal of Contemporary Asia, 42, 3, 359-377.
Cartwright, Nancy, 2010. 'What Are Randomised Controlled Trials Good For?', Philosophical Studies, 147, 1, 59–70.
Deaton, Angus and Nancy Cartwright, 2018. 'Understanding and misunderstanding randomized controlled trials', Social Science and Medicine, 210, 2–21.
Heckman, James and Jeffrey Smith, 1995. 'Assessing the Case for Social Experiments,' Journal of Economic Perspectives, 9, 2, pp. 85-110.
Kremer, Michael, Gautam Rao and Frank Schilbach, 'Behavioral Development Economics', in B. Douglas Bernheim, Stefano DellaVigna, David Laibson, eds, Handbook of Behavioral Economics: Foundations and Applications, 2, 345-458.
Mills, Anne, et al. 1995. 'Financing mechanisms for village activities in The Gambia and their implications for financing insecticide for bed net impregnation', Journal of Tropical Medicine and Hygiene, 97, 6, 325-32.
Ravallion, Martin, 2009. 'Should the Randomistas Rule?', Economists’ Voice, 6, 2, 1-5.
Ravallion, Martin, 2012. 'Fighting Poverty One Experiment at a Time: A Review of Abhijit Banerjee and Esther Duflo's Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty', Journal of Economic Literature, 50, 1, 103-114.
Ravallion, Martin, 2018. 'Should the Randomistas (Continue To) Rule?', Working Paper 492, Center for Global Development, New York (January 2019 update).
Ravallion 2020. 'Highly Prized Experiments', World Development, 127, 2020.
Reddy, Sanjay G. 2012. 'Randomise This! On Poor Economics', Review of Agrarian Studies, 2, 2.
Rodrik, Dani, 2009. 'The New Development Economics: We Shall Experiment, but How Shall We Learn?, in Jessica Cohen and William Easterly, eds, What Works in Development? Thinking Big and Thinking Small, Brookings Institution Press, Washington DC.
Thomas, Duncan, Elizabeth Frankenberg, Jed Friedman, et al., 2004. 'Causal Effect of Health on Labor Market Outcomes: Evidence from a Random Assignment Iron Supplementation Intervention', mimeo (available at https://escholarship.org/uc/item/1h66k92r).