63295 ResearchDigest World Bank VOLUME 5 ✬ NUMBER 1 ✬ FALL 2010 Charting New Directions in Development Economics Development economics needs to in the quantitative measurement of IN THIS ISSUE return to tackling the big questions poverty, and today the Bank is hailed Charting New Directions in Development of economic transformation and as the leading institution globally Economics … page 1 for data and research on develop- poverty elimination Is development economics in a position to ment. Research and data have long tackle the most urgent issues of the day? been essential elements of the Bank’s D evelopment economics, both country programs and its contribu- FOCUS within and outside the World tions to global public goods. Thanks Agriculture for Development, Bank, faces many challenges in no small part to the quality and Revisited … page 2 in its quest to inform policy mak- relevance of its research, the Bank now In the aftermath of the recent food and ers. Much has been learned about has a more pragmatic and empirical financial crises, a conference revisited the development from research and approach to policy making than ever issues addressed by the influential World operational experience. And develop- before. This is to be encouraged in Development Report 2008 ment data are more available and moving forward. Too Big to Fail or Too Big to Save?… page 3 accessible than in the past. But some But development thinking is in a “Too big to fail” has become a key issue in observers are asking whether develop- state of flux and uncertainty. It is time financial regulation. But have some banks ment economics in its present form to reconsider both the Bank’s research become too big to save? can tackle the most urgent issues priorities and how it does research. of the day. Even before the current Despite clear successes in many areas, Bank Runs in Emerging global crisis (which few economists concerns have been raised about the Economies … page 4 had forecast), there was a sense that relevance of some aspects of current While there has been much effort to the course of development economics development research. There are many strengthen banks in emerging economies, needed rethinking. A new paper pro- challenges ahead in improving devel- macroeconomic risk warrants more attention duced by the staff of the World Bank’s opment data, including expanding cov- Development Economics Senior Vice erage and improving internal consis- Closer Scrutiny of Mashup Indices … page 5 Presidency (affectionately known as tency. Too little attention has been giv- DEC) provides an overview of the en to the need for differentiated policy Composite indices of development are history of development research at recommendations for countries at popular. But should they be used to inform policy making? the Bank and points to future new different stages of development. And directions in both what it researches there is too little research on develop- Ethnic Discrimination and Political and how it researches. ment strategy, including on the poli- Competition in Sri Lanka … page 6 The World Bank has been at the cies and mechanisms that have been In ethnically heterogeneous societies, can forefront of development policy re- used by some fast-growing countries an electoral system based on proportional search throughout its history. Some of to accelerate structural transformation. representation help reduce ethnic the founding fathers of development Moreover, there is concern that discrimination? economics, such as Albert Hirschman much present-day research, both micro and Paul Rosenstein-Rodan, worked and macro, is too narrowly focused to Interactions between Formal and Informal for the Bank in the early 1950s. provide the kinds of insights needed to Institutions … page 7 Empirical research began in earnest help bring a large and lasting reduction New research looks at the implications of an in 1971 under the leadership of its in global poverty. Much emphasis is unreliable production process for formal and first chief economist, Hollis Chenery. informal institutions There were some important achieve- ments along the way, particularly (continued on page 8) 2 World Bank ResearchDigest FOCUS Agriculture for Development, Revisited The food, financial, and climate long-distance migration, consolidation rush of investments in large corporate crises have underscored the of land in larger farms requires vigor- farms in Argentina, Brazil, Indonesia, importance of agriculture and of its ous employment creation in the rural Kazakhstan, and Malaysia. As Klaus nonfarm economy, mostly in secondary Deininger and Derek Byerlee showed, role in development towns and cities. new sources of economies of scale The growth of employment in these have emerged. They originate in new M uch has changed since the towns is linked to agriculture but technologies, contracts with super- World Bank’s World Development also, importantly, to urban growth. market chains, public-private partner- Report 2008: Agriculture for Peter Lanjouw finds that employment ships for infrastructure, and access to Development was published in 2007. growth in the rural nonfarm economy international finance and reinsurance. With the food, financial, and cli- is strongly correlated with growth in The advantage of large farms has mate crises, increased exposure to neighboring urban centers. This sug- come from market failures and policy shocks—and its effects on food inse- gests a need to focus much more on distortions, creating biases against curity and vulnerability to poverty— promoting the small towns and cities smallholders. Eliminating these has become a dominant concern. of rural India. Doing so would allow biases—for example, through producer The World Bank and the Center of both land concentration and poverty organizations—can sometimes restore Evaluation for Global Action orga- reduction in rural areas. the competitiveness of family farms. nized a conference at the University of The link between price volatility and After World Development Report 2008 California, Berkeley, on October 1–2 food security was another topic dis- was published, donors signaled their to revisit the issues addressed by the cussed. Overall, price transmission be- agreement that agriculture deserved report. tween domestic and international mar- more investment and that food se- A key finding presented in the kets has been surprisingly low, uneven, curity and hunger needed more at- report was that GDP growth originat- and difficult to explain. Will Martin, in tention. This was manifest at the G-8 ing in agriculture was two to three a paper written with Kym Anderson, summit in L’Aquila, Italy, in 2009, times as effective in increasing the showed that this has implications where participants pledged $22 billion consumption spending of the poor as both in insulating markets from inter- for agriculture. If these commitments GDP growth originating in the rest of national price movements (as desired are honored, would we know how to the economy. This may be changing by politicians) and in reducing the spend the money effectively? How to as populations become increasingly stabilization of domestic prices when best use agriculture for development urbanized. Using data for India, Martin there are local food production shocks is a question requiring rigorous experi- Ravallion showed that urban growth (likely to increase as climate change mentation and learning. Creative new has become more effective in reducing accelerates). approaches to rural development have overall poverty. Urban growth also ap- Beggar-thy-neighbor policies have been introduced. Two were explored at pears to be just as effective in reduc- not been seriously addressed, as the conference. Jo Swinnen discussed ing rural poverty as it is in reducing there is virtually no international co- contracting between smallholders and urban poverty. ordination in managing responses to commercial partners in high-value The implication for rural pov- food price shocks. Political pressures food chains. Jenny Aker discussed the erty reduction is important. Mark toward food self-sufficiency and anti- use of information and communication Rosenzweig, in a paper written with trade positions may undermine rather technology in agricultural extension. Andrew Foster, presented new results than support the quest for a solution. Revisiting the state of world agri- for India showing that increasing farm Brian Wright showed that universal culture for development three years size is key to improving incomes in food price subsidies, as pursued in after the report’s publication reveals agriculture. Expanding farm size al- Middle Eastern and North African tremendous opportunities, important lows the use of mechanization that countries, have large fiscal costs for new advances, and a hugely incom- has indivisibilities (with differences in limited benefits. And land purchases plete task requiring urgent progress. access to credit favoring large farms abroad to achieve food security may Much remains to be learned about how over small ones), implying increasing not be immune to political interven- to use agriculture for development, returns to scale and higher profit- tions in the midst of crises. conceptually and in practice. ability per hectare. Today most farms Rapidly growing demand for food in India are too small to exploit the and feed in emerging economies productivity and cost-saving advan- and rising world market prices have tages of mechanization. But with con- created attractive investment op- Papers, presentations, and the conference program tinued population growth and limited portunities in agriculture, leading to a can be found at http://cega.berkeley.edu/agfordev. World Bank Research Digest 3 Too Big to Fail or Too Big to Save? Systemically large banks may face shareholders as discounted in bank strong incentives to reduce their size private incentives to downsize stock prices. They also distinguish relative to the national economy. between systemically important and Indeed, a smaller percentage of banks I n financial regulation, “too big smaller banks. In a parallel fashion, were systemically important (relative to fail” has become a key issue. they consider the impact of govern- to GDP) in 2008 than in the two previ- Indeed, during the recent crisis ment finances on expected losses on ous years (figure 1). many financial institutions received banks’ liabilities, as reflected in five- The problem of “too big to save” subsidies precisely because they were year credit default swap (CDS) spreads. facing systemically large banks in fis- deemed too big to fail by policy mak- Specifically, they consider bank valu- cally strapped countries is likely to ers. The expectation that large institu- ation in 1991–2008, with 717 publicly change the structure of the interna- tions will be bailed out by taxpayers listed banks in 34 countries in 2008, tional banking system. Banks in all any time they get into trouble makes and CDS spreads in 2001–08, with 59 financial systems will face pressure the job of regulators all the more diffi- banks in 20 countries in 2008. to deleverage in order to reduce risks cult. After all, if someone else will pay Overall, the authors find that while for themselves and for the financial for the downside risks, institutions are systemically large banks may benefit safety net. But systemically large banks likely to take on more risk and get into more than smaller banks from taking in fiscally constrained countries will trouble more often. The many implicit on more risk, they can also suffer from have particularly strong incentives to and explicit benefits that govern- being in a country that runs large gov- downsize in order to be able to rely on ments are willing to extend to large ernment deficits at a time of financial the financial safety net in the future. institutions make reaching too-big-to- crisis. This makes the net benefit of The evidence suggests that downsiz- fail status a goal in itself for financial systemic size ambiguous. The results ing should increase bank valuation. institutions. Thus all the proposed also suggest that some banks may The downsizing that occurred in 2008 legislation to tax away some of these have grown beyond the size that maxi- thus may have been driven in part by a benefits. mizes their implicit subsidy from the desire to increase stock market valua- But could it be that some banks financial safety net. Such banks can tion in the face of the too-big-to-save have become too big to save? This is increase shareholder value by downsiz- effect—even if losses and difficulties a valid question, particularly for small ing or splitting up. For the overall sam- in raising equity and other types of countries or those with deteriorating ple, the results suggest that the share capital at a time of financial crisis also public finances. The prime example is price of systemically important banks played a role. Iceland, where the banking system’s li- is discounted 22.3 percent on average abilities rose to about nine times GDP because of systemic size, providing (continued on page 7) at the end of 2007, before its spec- tacular collapse in 2008. By the end of Figure 1. Systemically Large Banks as a Percentage of Publicly Listed Banks 2008 the liabilities of publicly listed in Selected Countries, 1991–2008 banks in Switzerland had reached 12 6.3 times GDP, while the liabilities of those in the United Kingdom had 10 risen to 5.5 times GDP. Big0.1 In a recent paper Demirgüç-Kunt 8 and Huizinga investigate whether market valuation of banks is sensitive Percent 6 to government indebtedness and defi- cits. Markets may doubt the ability of financially strapped countries to save 4 Big0.5 Big0.25 their largest banks. At the very least, governments in this position may be Big1.0 2 forced to resolve bank failures in a relatively cheap way, implying large 0 losses to bank creditors. 91 01 00 93 96 03 06 92 98 99 94 04 02 08 07 95 05 97 The authors investigate the impact 19 20 20 19 19 20 20 19 19 19 19 20 20 20 20 19 20 19 of a country’s public finances, in the Note: The figure shows the percentages of banks with a ratio of liabilities to GDP exceeding various thresholds for 34 countries. Big0.1 shows the percentage with a ratio exceeding 0.1, Big0.25 the percentage with a ratio form of both government debt and exceeding 0.25, Big0.5 the percentage with a ratio exceeding 0.5, and Big1.0 the percentage with a ratio deficits, on expected returns to bank exceeding 1.0. 4 World Bank ResearchDigest Bank Runs in Emerging Economies Recent episodes of bank runs have or through a gradual deterioration of macroeconomic indicators at the ex- reignited interest in understanding traditional bank-specific characteris- pense of bank-specific characteristics. what drives the behavior of tics. An example of exposure to macro- The findings yield important les- economic risk not captured by typical sons for the policy debate. First, the depositors bank-specific measures of fundamen- quest for market discipline, embed- tals is foreign currency lending. In fi- ded in pillar 3 of Basel II, moves in the C urrent research on how deposi- nancially dollarized economies, where right direction by addressing bank su- tors behave mainly emphasizes banks match their foreign currency pervisors’ limitations in enforcing pru- their response to bank-specific funding with foreign currency lend- dential regulation. But it faces serious characteristics—what is known in ing, exchange rate devaluations may shortcomings—particularly in emerg- banking as market discipline. For significantly increase credit risk as they ing economies, where market reaction example, depositors respond to in- impinge on the repayment capacity of is often driven by macroeconomic creases in bank risk (bank-specific unhedged foreign currency borrowers conditions largely beyond the control characteristics) by demanding higher (for example, those in the nontradable of bank managers. This raises ques- interest rates on their deposits or by sector). A gradual deterioration in tions about the perception of market withdrawing their funds, penalizing traditional bank-specific indicators due sensitivity as a disciplining device. managers for excessive risk taking. to macroeconomic risk might occur, From a prudential perspective, mar- Market discipline, if present, tends for example, as a result of an increase ket responses to idiosyncratic factors to lower the probability of individual in sovereign risk, which will adversely should be distinguished from market bank failures and the incidence of affect the return on assets for banks responses to macroeconomic factors. banking crises and lead to a healthier holding government paper. Second, the finding that macroeco- banking sector as a whole. To document the importance of nomic risk affects banks suggests that Some studies discuss the pos- macroeconomic risk for the behavior supervisors could play an active role in sibility that generalized deposit runs of depositors, the authors focus on reducing exposures through prudential can be random events or triggered by the episodes of bank runs in Argentina regulation. For example, liquid asset contagion from weak banks and thus and Uruguay in 2000–02. The analysis and provisioning requirements could unrelated to bank-specific character- of these two crises leads to several take into account the denomination istics. A recent paper by Levy Yeyati, findings. First, it shows that macro- of bank loans. Simple measures to ad- Martínez Pería, and Schmukler shows economic shocks can quickly cause dress macroeconomic exposures are that macroeconomic risk can also ex- the collapse of banking systems, typically absent from standard pruden- ert a significant—and, in many cases, while low-frequency bank-specific tial best practices. In addition, regula- decisive—influence on the evolution factors do not explain these events. tors need to be acutely aware of struc- of bank deposits. These episodes are not random or tural susceptibility to macroeconomic Macroeconomic risk can influence driven by contagion across banks. shocks and cannot expect price signals depositor actions both regardless Macroeconomic factors that are largely to alert them to those susceptibilities. of and through bank-specific char- irrelevant in explaining depositor In recent years much effort has acteristics. The effects of macroeco- behavior during tranquil times become been put into strengthening banks nomic factors beyond bank-specific the main driver of market response in emerging economies. Surprisingly, characteristics can take place when during crisis episodes, even after the there has been relatively less empha- worsening macroeconomic condi- analysis controls for standard bank- sis placed on limiting the exposure tions directly threaten the value of specific traits. to macroeconomic risk that can ulti- market participants’ assets (such as Second, the analysis shows that mately bring down the entire system. bank deposits). A classic example of even though bank-specific exposure The authors help fill this gap by show- direct macroeconomic effects is cur- to macroeconomic risk is not factored ing empirically that a more balanced rency risk: depositors might flee from into banks’ funding costs in the precri- effort is indeed warranted to enhance domestic banks, regardless of the sis period, deposit withdrawals during the stability and resilience of the bank- fundamentals of individual banks, if the crises are positively associated ing sector where macroeconomic risk convertibility to a foreign currency is with banks’ exposure to exchange prevails. not an option. rate risk. Finally, the authors illustrate The effects of macroeconomic risk how the informational content of Eduardo Levy Yeyati, Maria Soledad Martínez on depositor behavior through bank- bank-specific characteristics dete- Pería, and Sergio Schmukler. 2010. “Market specific characteristics can take place riorates relative to macroeconomic Discipline under Macroeconomic Risk: Evidence either through exposure to macroeco- factors as aggregate risk mounts, from Bank Runs in Emerging Economies.” Jour- nomic risk (not typically captured by which explains why risk-aware deposi- nal of Money, Credit, and Banking 42 (4): the most frequently used indicators) tors may increase their response to 585–614. World Bank Research Digest 5 Closer Scrutiny of Mashup Indices Users should look closely at the where their country stands. Mashup flexibility into the Web sites so that properties and robustness of some indices have often increased public users could customize the index with popular composite indices of awareness about important develop- their preferred weights and see what ment issues. But these indices need difference it makes. development closer scrutiny. The following ques- How should the index be used by policy tions should be asked about any new makers? Given the real constraints that A host of indicators are used to mashup index. countries face, it is not credible that track development. The World What is the index measuring? The fact any one of these indices is a sufficient Bank’s annual World Development that the target concept is unobserved statistic for country performance. We Indicators presents hundreds of such does not mean we cannot define it and might well rank the countries very indicators. The United Nations postulate what properties we would differently if we took account of their Millennium Development Goals are like its measure to have. Yet this is not stage of economic development. defined using a long list of indicators. common in the industry of mashup in- Policy makers might be better ad- For some of the composite indi- dices. The frequent lack of conceptual vised to use the component measures ces in use, economic theory provides clarity about what exactly one is trying most appropriate to each policy instru- some useful clues on how the index to measure makes it hard to judge the ment rather than the mashup index. should be constructed. This is not so practical choices made about what While many things affect your personal for another type of composite index preexisting indicators get used in the health, you would not want your doc- that is becoming popular. For these in- composite. tor to base your checkup on a compos- dices neither the menu of the primary What tradeoffs are embedded in the index? ite index. Similarly, a mashup index of series to be aggregated nor the aggre- We need to know the tradeoffs built all those dials on your car’s dashboard gation function is predetermined from into the index if it is to be properly may fail to reveal that you are about to theory and practice; both are instead assessed and used. At one level the run out of fuel. “moving parts”—key decision variables weights in most mashup indices are Arguably, mashup indices exist be- that the analyst is essentially free to explicit. Common practice is to identify cause theory (and not just economic choose. Borrowing from Web jargon, a set of component variables, group theory) has given too little attention we can dub these “mashup indices.” these in some way, and attach equal to the full range of measurement prob- Probably the most famous example weight to these groups. But little or no lems faced in assessing development is the Human Development Index, attention is given to what the implied outcomes. Theory needs to catch up. published each year in the Human tradeoffs are in the space of the prima- Thankfully, progress in develop- Development Report. This index is a ry dimensions being aggregated and ment does not need to wait for that composite of life expectancy, school- whether those weights are defensible. to happen. A composite index is not ing, and log real GDP per capita. In a Nor do the implied tradeoffs appear to essential for many of the purposes of similar spirit the Multidimensional have been taken into account in choos- evidence-based policy making. Indeed, Poverty Index, created for the 2010 ing the aggregation functions for most it may even distort policy making by edition of the Human Development mashup indices of development. encouraging policy makers to focus on Report, aggregates household-level How robust are the rankings? Theory readily observed proxies rather than deprivations in health, education, never delivers a complete specifica- the deeper characteristics of the econ- and living standards. Mashups have tion for measurement, so judgments omy and society that ultimately matter been devised for other dimensions of are required about one or more pa- for development outcomes. development, including governance rameters. There is also statistical im- (notably the Worldwide Governance precision about parameter estimates. Indicators), private enterprise regula- For these reasons it is widely recom- tion (most prominently the Ease of mended scientific practice to test the Doing Business Index), and the envi- robustness of the derived rankings. But ronment (such as the Environmental users of mashup indices are rarely told Performance Index). Probably the much about the uncertainties that ex- most ambitious example is Newsweek ist about the series chosen, the qual- magazine’s effort to identify the ity of the data, and their weights. Few “world’s best countries” using a com- rigorous robustness tests are provided. posite of many indicators, some of None of the Web sites for mashup in- them already mashups. dices make it easy for users to properly The country rankings based on assess the sensitivity of these indices Martin Ravallion. 2010. “Mashup Indices of De- such indices often attract media atten- to changing weights. Yet it would be velopment.” Policy Research Working Paper 5432, tion. People are naturally keen to see relatively easy to program the required World Bank, Washington, DC. 6 World Bank ResearchDigest Ethnic Discrimination and Political Competition in Sri Lanka Political competition can lessen Samurdhi food stamps in politically for making ethnically divided societ- ethnic discrimination in the important areas. ies more inclusive. They suggest that allocation of public expenditure The paper uses data on divisional- political competition in ethnically level voting patterns in the Sri Lankan heterogeneous settings may serve as a general elections of 1994 as well as check on ethnic bias in the allocation T he discourse among economists data from the 1999/2000 Sri Lankan of public resources. on ethnic discrimination has Integrated Household Survey to test The results also support models on focused largely on explanations whether there was ethnic discrimina- pork barrel politics that suggest that for the phenomenon and for its tion in the allocation of Samurdhi food incumbent governments “purchase” consequences—and much less on stamps and to what extent political votes by allocating discretionary trans- how to address this complex problem. competition undermined the ethnic fers to population groups considered But leading political theorists have of- discrimination effects in their alloca- to represent the swing or core elector- fered a plausible solution, proposing tion. The second effect is teased out by ate. These results suggest a fundamen- proportional representation for the using two measures of political com- tal tension in government-targeted electoral system in divided societies. petition: the levels of core support and transfer programs, where gains in Under proportional representation, it marginal support for the incumbent social protection have to be weighed is believed, strong electoral incentives at the divisional level. Because one against losses caused by politicians force politicians in heterogeneous of the manifestations of ethnic dis- who may prefer to favor undeserving societies to seek votes outside their crimination is lack of access to public groups in the allocation of transfers so own ethnic group to win elections resources, such an analysis represents as to benefit politically. For example, and maintain coalitions. The need to an empirical test of the hypothesis the spread of Samurdhi consumption engage in such “vote pooling” is what that democratic competition character- transfers across the Sri Lankan popu- encourages politicians to accommo- ized by incentives provided by elec- lation, if driven by political motives, date the interests of ethnic minorities toral arithmetic and the composition could very well divert scarce resources and thus mitigate discriminatory prac- of constituencies can lessen ethnic away from the neediest groups. But tices. An underlying implication is discrimination. this problem is not unique to ethni- that politicians may seek to avoid dis- The results support the hypothesis cally heterogeneous societies. criminating against minority groups that political competition helps to The results of the paper are consis- when such groups are an important lessen discrimination toward ethnic tent with the predictions of research part of their party’s current core sup- minorities. The paper finds evidence that certain types of institutions are port base or when the overall elector- that Sri Lankan Tamil households, conducive to ethnic harmony—and al system is so competitive that there which are generally less likely to re- that democracy is one of them. That is a high probability that the govern- ceive government transfers than the said, there is always scope for manipu- ing party will need minority support majority Sinhalese population, have lating the electoral system, and studies to hold onto power. a higher chance of receiving such have yet to identify any variant of the Most of the theoretical observa- transfers when they live in areas where proportional representation system tions on political competition and political competition is tight or sup- that has succeeded in permanently interethnic cooperation are normative port for the incumbent political party minimizing ethnic divisions and dis- and have not been put to an empiri- is high. Political competition has a crimination. Each situation is complex, cal test, mainly because of the lack of greater effect in reducing discrimina- reflecting the constraints of the histori- systematic data on ethnic discrimina- tion when Tamil minorities form part of cal, political, and power dynamics in tion. Using data on the allocation of swing constituencies than when they which it is formed and the negotiating government food stamps in Sri Lanka, form part of the base support. skills of the parties involved. More re- a country with ethnic heterogeneity The analysis seeks to minimize search is needed to test the effective- and many decades of ethnic turmoil, (though not exclude) the possibility ness of different political approaches a recent paper by Sharif investigates of omitted-variable bias in the results that could be adopted to address eth- whether political competition affects by using a host of household- and nic discrimination. the probability of ethnic minorities’ community-level variables, including receiving government transfers. The community-level ethnic heterogene- basic electoral incentives model that ity, to control for differences between is tested in the paper applies the politically important areas and the rest theory of both vote pooling and pork of the country. Though not conclusive, Iffath A. Sharif. Forthcoming. “Does Political barrel politics to examine whether the results are consistent with other Competition Lessen Ethnic Discrimination? Evi- the ethnicity of the voter matters in research that considers the design of dence from Sri Lanka.” Journal of Develop- the allocation of the government’s political institutions to be a key lever ment Economics. World Bank Research Digest 7 Interactions between Formal and Informal Institutions The reliability of the production of firms to bribe because of the higher Once country and sector character- process is a physical channel rents involved and the incentives of istics are taken into account (through through which development can consumers to connect with one an- fixed effects), the authors observe that other to identify poorly performing in all regions the reliability index has influence institutions firms. At the same time, however, un- a statistically significant relationship observed productivity shocks keep the with both membership in business I n many developing and transition equilibrium price higher than marginal associations and corruption. This con- economies informational and con- costs, to encourage firms to foster firms what the theory predicts: lower tracting constraints make it more high quality. In particular, the higher reliability of a market is associated difficult for enterprises to do busi- the frequency of productivity shocks with a higher likelihood that a firm is a ness. These constraints are perceived (a typical feature of low- and middle- member of a business association and to stem mainly from dysfunctional income countries), the higher the price with higher corruption. formal enforcement institutions, must be relative to the actual costs of such as courts. To deal with these production. This provides a new ex- constraints, enterprises use informal planation of why some goods, even in Amrita Dhillon and Jamele Rigolini. Forthcoming. mechanisms of contract enforcement a competitive setting, may be priced “Development and the Interaction of Enforcement based on reputation, networks, and higher in low- and middle-income Institutions.” Journal of Public Economics. relationships. There is ample evi- countries. dence of this from around the devel- Second, the theory suggests a oping and developed world. Existing physical channel through which devel- studies consider formal and informal opment can influence institutions: the institutions to be full substitutes. But reliability of the production process. this is unrealistic, since these two An unreliable production process af- modes of governance are extremes. fects institutions and the welfare of How formal and informal institutions consumers in several ways. Beyond (continued from page 3) interact remains a central question. leading to higher equilibrium prices, it A recent paper by Dhillon and provides firms with greater incentives Too Big to Fail or Too Big to Rigolini addresses this open question to bribe if a case is brought to court Save? by developing a new theory of insti- because higher rents are involved. In tutional interactions in which formal addition, it directs consumers toward and informal institutions coexist and informal networks and personal con- There is an obvious policy inter- influence one another. The authors de- nections in order to identify poorly est in reducing bank size to below the velop a theoretical framework in which performing firms. point where banks’ national contingent unobserved firm-level shocks, related Preliminary evidence suggests that liabilities are so large that there are to the level of development, imply such a channel could play a signifi- doubts about governments’ ability the production of poor-quality goods. cant role. The authors use World Bank to stabilize the banking system. This Such shocks generate a moral hazard Investment Climate Survey data from also at least partially explains the cur- problem because consumers cannot six regions to study the determinants rent proposals to limit bank size or determine whether poor quality stems of firms’ membership in business as- tax systemically large banks. But that from a shock or from firms cheating sociations (a proxy for informal con- the percentage of systematically large and intentionally producing goods of nectedness) and of their perception banks had already declined in 2008 low quality. To cope with this problem, of corruption. A firm is considered a even without additional regulation and consumers rely on two enforcement member of a business association only taxation may reflect private incentives mechanisms: an informal one, based if it belongs to the association and rec- to downsize in the face of a too-big- on reputation and networks, which is ognizes its ability to resolve disputes to-save effect in fiscally constrained enhanced through consumers’ invest- and provide information on domestic countries. Additional regulation or ment in being “connected”; and a for- product markets. The authors create a taxation aimed at very large banks may mal mechanism, which acts through corruption variable that accounts for strengthen this trend. legal enforcement (but can be weak- all the firms that identify corruption as ened by firms through bribes). an important constraint to business. This theory makes two contribu- In addition, using the principal com- Aslı Demirgüç-Kunt and Harry Huizinga. 2010. tions. First, it highlights the impor- ponents method, they construct a reli- “Are Banks Too Big to Fail or Too Big to Save? tance of market prices as a channel of ability index based on whether firms International Evidence from Equity Prices and institutional interaction. High equilib- identify electricity shortages, transport, CDS Spreads.” Policy Research Working Paper rium prices affect both the incentives and skills as important constraints. 5360, World Bank, Washington, DC. 8 World Bank ResearchDigest (continued from page 1) given to internal validity—establishing • Assessing the effectiveness and Recent Policy Research that the conclusions about the setting results of development efforts, includ- Working Papers follow from the data. But research has ing external assistance. not brought the same rigor to address- The multipolar world that is emerg- 5405 Economic Freedom, Human Rights, and the ing external validity—determining ing requires a new, multipolar ap- Returns to Human Capital: An Evaluation of whether findings allow the drawing of proach to knowledge. World Bank the Schultz Hypothesis Elizabeth M. King, Claudio E. Montenegro, lessons for other settings, including researchers have always worked with and Peter F. Orazem scaling up. external researchers. For example, 5412 Improving Wastewater Use in Agriculture: An Emerging Priority A more open and strategic ap- consultants coauthoring research Susanne M. Scheierling, Carl Bartone, proach to research is needed—an publications with Bank staff during Duncan D. Mara, and Pay Drechsel 5414 Explaining Variation in Child Labor Statistics approach firmly grounded in the key 1998–2005 came from 66 developing Andrew Dillon, Elena Bardasi, Kathleen knowledge gaps for development countries and 25 developed countries. Beegle, and Pieter Serneels policy emerging from the experiences But the Bank should collaborate more 5417 Global Value Chains in the Electronics Industry: Was the Crisis a Window of of developing countries, including the intensively with developing-country Opportunity for Developing Countries? Timothy J. Sturgeon and Momoko Kawakami questions that policy makers in those researchers and institutes. 5423 Political Economy of Infrastructure Spending countries ask. Development econom- Moreover, greater emphasis must in India ics needs to return to tackling the big be given to producing the data and Stuti Khemani 5424 Distributions in Motion: Economic Growth, questions of how to sustain economic analytic tools for others to do the re- Inequality, and Poverty Dynamics growth and how to eliminate pov- search themselves and providing open Francisco H. G. Ferreira 5425 How and Why Does History Matter for erty, but to do so with all the power access to those tools. The open data Development Policy? that data and analytic tools can now initiative recently launched by the Michael Woolcock, Simon Szreter, and Vijayendra Rao provide. World Bank needs to be extended to 5444 The Impact of the Financial Crisis on New Firm Four major sets of problems merit open knowledge. This will better in- Registration Leora Klapper and Inessa Love high priority for future research on de- form development policy debates and 5445 Payment Systems, Inside Money and velopment policy: allow deeper engagement with those Financial Intermediation • Achieving efficient economic who have a direct stake in the out- Ouarda Merrouche and Erlend Nier 5446 Islamic vs. Conventional Banking: Business transformation, which also means comes of those debates. Model, Efficiency and Stability the ability to tailor policy advice to Thorsten Beck, Aslı Demirgüç-Kunt, and Ouarda Merrouche different stages of development. 5450 Subnational Taxation in Developing • Broadening opportunities to en- Countries: A Review of the Literature Richard M. Bird sure that all individuals can participate 5455 Dealing with Politics for Money and Power in in the benefits of, and contribute to, World Bank, Development Economics Senior Vice Infrastructure Daniel Benitez, Antonio Estache, and Tina this transformation. Presidency. 2010. “Research for Development: A Søreide • Dealing with emerging new World Bank Perspective on Future Directions for 5456 The Co-movement of Asset Returns and the global challenges related to risk and Micro-Macro Focus of Prudential Oversight Research.” Policy Research Working Paper 5437, Giovanni Majnoni vulnerability. World Bank, Washington, DC. Working Papers can be downloaded at http://econ.worldbank.org To download the World Bank Research E-Newsletter, go to Data & Research at http://www.worldbank.org The World Bank Research Digest is a quarterly publica- The Research Digest is financed by the Bank’s Editorial Committee: Jean-Jacques Dethier (managing tion disseminating findings of World Bank research. Research Committee and managed by DECRS, the editor) and Aslı Demirgüç-Kunt. Editor: Alison Strong; The views and interpretations in the articles are those research support unit of the Development Economics production: Roula Yazigi. For information or free of the authors and do not necessarily represent the Senior Vice Presidency (DEC). The Research Digest is subscriptions, send email to research@worldbank.org views of the World Bank, its Executive Directors, or the not copyrighted and may be reproduced with appropri- or visit http://econ.worldbank.org/research_digest. countries they represent. ate source attribution. The World Bank 1818 H Street, NW Washington, DC 20433, USA Printed on Recycled Paper