Thomas Edison, the intrepid American inventor who brought to the world the lightbulb and held more than a thousand patents, had a simple rule for recruiting his engineers. He always invited the short-listed candidates (whom he assumed to all be technically capable of doing the job at hand) for lunch and carefully observed their behavior. 

His objective was not to check their table manners, but to deduce their decision-making process from their most anodyne acts. A key indicator was his guests’ propensity to add salt and pepper to their dishes without first tasting them. That very common and often unconscious gesture would disqualify any prospective candidate: it revealed a tendency to blindly act on one’s instincts and to decide without checking the evidence.

Many researchers and policymakers working today on institutional development in developing countries are guilty of the Pavlovian behavior despised by Edison. They assume that, by simply looking at the current state of institutional development in industrialized countries, they know precisely what it means and how it can be measured. They mechanically compare political, administrative and financial institutions of countries regardless of their economic development levels. They naturally find gaps between poor and rich countries, and derive from those a generic reform agenda not based on evidence.

They also neglect lessons from the history of highly industrialized economies, all of which started their development success stories with suboptimal political, administrative and financial institutions. The broad (and somewhat abstract) intuition on the need to improve governance in all countries has strong moral and theoretical foundations: it is the right thing to do to sustain growth, ensure shared prosperity and build social trust and stable societies. But the conventional wisdom that low-income countries should therefore start their development with the governance institutions of high-income countries is both a non sequitur and a historical fallacy.

This chapter acknowledges that institutional development problems are indeed major impediments to economic growth. But contrary to conventional wisdom, it argues that they are often simply correlated with the level of economic development — and correlation is not causation. Seen from that perspective, the well-known weaknesses in the governance and financial sectors of many poor countries today often reflect their low level of development and the results of failed state interventions and distortions originating from erroneous economic development strategies.

Instead of posing first-world governance and financial institutions as the main prescription for sustained growth in third-world countries, economists should design policies that offer the maximum likelihood of success because they are consistent with comparative advantage and existing firm structure, while minimizing opportunities for rent-seeking and state capture. The dynamic development of competitive firms and industries will eventually lead to institutional development.

The Mystery of Governance

Institutional development is generally acknowledged to be the reflection or result of “good governance.” Yet as A. Premchand, an economist at the International Monetary Fund once observed, governance is like obscenity “because it is difficult to define.”

Perhaps the most comprehensive and authoritative intellectual source on the subject are the Worldwide Governance Indicators, produced in conjunction with the World Bank and widely respected in policy circles. They define governance as:

The traditions and institutions by which authority in a country is exercised. This includes the process by which governments are selected, monitored and replaced; the capacity of the government to effectively formulate and implement sound policies; and the respect of citizens and the state for the institutions that govern economic and social interactions among them.

They identify six dimensions of governance: government effectiveness, regulatory quality, rule of law, control of corruption, voice and accountability, and political stability and absence of violence. That conceptual framework is then given empirical life through the use of data produced by a variety of think tanks, survey institutes, international organizations, nongovernmental organizations and private-sector firms. The WGI are therefore aggregate indices that combine the views of a large number of respondents, including those representing enterprises, citizens and experts in industrialized and developing countries.

Such a valiant effort to give meaning to the complex notion of governance is certainly respectable. But the theoretical and philosophical underpinnings of the WGI are highly questionable. First, assessing the quality of the traditions and institutions by which authority in a country is exercised is bound to be a subjective exercise. It is therefore susceptible to being a reflection of ethnocentric — if not paternalistic — perspectives. There is no reason to believe that such an assessment should be performed uniformly in China, Alaska and Zanzibar.

There will always be those who claim that all human societies share the same goals and have adopted global standards and broad principles of good governance embodied in internationally agreed covenants. Yet there will also always be the perception that these global standards are actually evidence of the Westernization of human values under the pretense of “universality.” Both camps have some intellectual legitimacy. The reality that some who reject the good-governance agenda as a hidden attempt to Westernize the world may be defenders of authoritarian practices hidden behind the claim of cultural relativism does not necessarily invalidate all their arguments.

The WGI and other indicators of good governance or democratization do not really help escape the universalist-versus-relativist impasse. Moreover, even if one could come up with an ingredient list that satisfies both the universalists and the relativists, the belief that good governance can be captured quantitatively and measured through surveys will remain subject to debate.

Behavioral economics shows that people often err when asked to identify the constraints that affect even their most important activities. Econometric analyses show, for instance, that popular survey results such as the World Bank’s Doing Business indicators do not correlate well with the actual constraints on private-sector performance. In other words, even the most successful businesspeople in the world generally fail to intuitively identify the real obstacles to productivity growth and enterprise development. If that is the case, how confident can one be about perceptions of others’ well-being?

In fact, there are fundamental discrepancies between indicators of what is perceived as good governance and indicators of actual economic performance. These discrepancies also reveal fundamental issues of subjectivism and ethnocentrism that are reminiscent of the “orientalism” analyzed by Edward Said, the Palestinian-American who pioneered the field of post-colonial studies.

A good illustration of the problem can be found in countries’ corruption-scale rankings, always one of the key pillars of the good governance agenda. Transparency International surveys basically show that the world is divided into two categories: the highly corrupt countries and the least corrupt, with a “good-governance” Western world surrounded by a “bad-governance” non-Western world. Transparency International is a reputable organization that does good work. But its surveys, which display a Manichean view of the world, are deeply disturbing.

Given that the fight against corruption is an important part of the good-governance agenda, it is perplexing that corruption has been prevalent throughout human history and still exists, often on a wide scale, in high-income countries. In recent years, France gave its former president Jacques Chirac a two-year suspended prison sentence for diverting public funds and abusing public trust. In the United States, four of the past seven Illinois governors were convicted and imprisoned — among them, Rod Blagojevich, who was convicted of numerous corruption charges including for trying to “sell” President Barack Obama’s former Senate seat to the highest bidder. In Japan, many high-ranking government officials have been forced out of office throughout the postwar period amid corruption scandals. The problem has extended well beyond the political sphere and into a bureaucracy often considered one of the better managed in the world.

The typical response to the unflattering truth is to argue that high-income countries are “less” corrupt than others and their institutions are “stronger.” But those arguments are hard to empirically validate. First, it is difficult to rigorously define what corruption means every time and everywhere, and to compare it across time and place. Many open transactions between lobbyists and policymakers in the United States, for example, would be considered corruption in other places in the world. Consider, too, that aggressive prosecution of corruption is taking place in many low-income countries, yet in those countries such legal actions are paradoxically considered further evidence of terrible governance. A case in point is that of Cameroon, where dozens of politicians and civil servants at the highest levels of power (including a former prime minister) have been convicted for embezzling public funds. Yet few analysts would consider Cameroon a good-governance country. To the contrary, the more senior government officials sent to jail, the more experts are convinced that Cameroon is profoundly corrupt.

Defining Corruption

The first clue to the inextricable difficulties of corruption analytics is the vagueness of the definitions that one can find in the technical literature. The most commonly used definition is the one by Andrei Shleifer and Robert Vishny, who define government corruption as:

The sale by government officials of government property for personal gain. For example, government officials often collect bribes for providing permits and licenses, for giving passage through customs or for prohibiting the entry of competitors. In these cases they charge personally for goods that the state officially owns. In most cases the goods that the government officials sell are not demanded for their own sake, but rather enable private agents to pursue economic activity they could not pursue otherwise. Licenses, permits, passports and visas are needed to comply with laws and regulations that restrict private economic activity. Insofar as government officials have discretion over the provision of these goods, they can collect bribes from private agents.

The obvious question that this well-established definition raises is that of legality. What if there are no laws in place preventing government officials from making excessive or arbitrary use of their discretionary power? Does corruption intrinsically imply illegality? If it does, then the logical inference from the definition is that some practices may be considered “corruption” in some countries and not in others.

Corruption can be disaggregated along several dimensions. First, one must distinguish its prevalence, especially in large countries with decentralized political systems, where corruption can be widespread at the local government level, even if it is controlled at the central government level. Second, the purpose of improper actions characterized as corruption must also be taken into account. Bribes intended to influence the design and content of laws and regulations (state capture) must be differentiated from those intended to change or circumvent their implementation (administrative corruption). Third, there is a need to distinguish among the actor categories involved in various forms of corruption: when poor people are involved, it is often referred to as petty corruption as opposed to grand corruption, which involves high-level officials and political figures. Fourth, corruption may be of a different scale and nature depending on the administrative agency in which it takes place (schools, customs, health centers and so on).

Transparency International has chosen a more focused operational definition of the term: “the abuse of entrusted power for private gain.” The organization further differentiates between “according to rule” corruption and “against the rule” corruption. Facilitation payments, where a bribe is paid to receive preferential treatment for something that the bribe receiver is required to do by law, constitute the first. The second is a bribe paid to obtain services the bribe receiver is prohibited from providing.

While Transparency International’s definition of corruption is much clearer, it raises another series of problems. First, since bribe payments are not publicly recorded, it is virtually impossible to calculate their frequency or magnitude. Second, bribes do not always take monetary form — favors, presents, services and even threats and blackmail are just as common. These factors highlight other issues, such as the strength of the judicial system and its ability to effectively handle complaints at the lowest possible cost, the prevailing cultural and behavioral norms, and so on.

Corruption’s social costs are even less quantifiable. As Transparency International points out:

No one knows how much the loss of an energetic entrepreneur or an acclaimed scientist costs a country. Moreover, any estimated social costs in dollars would be inadequate to the task of measuring the human tragedy behind resignation, illiteracy or inadequate medical care.

In fact, much of the research on governance implicitly suggests that corruption can be observed only on a wide scale in countries under a certain income level. Although there is obviously recognition that corruption also occurs in high-income countries, it is treated as poor behavior by public officials or businesspeople who represent the exception and not the rule. These high-profile, headline-grabbing cases are considered outliers and therefore are either ignored or discounted by mainstream economic research. Nothing could be more misleading.

Defining and measuring corruption is a difficult task, not least because the definitions differ not only between countries but within them. For instance, while all of Japan is subject to one penal code, the United States has 50 different penal codes (one per state) as well as a national (federal) code. Moreover, Japan keeps detailed statistics on corrupt acts, whereas the United States has no centralized record-keeping for such acts.

Still, consider these statistics: between 1987 and 2006, U.S. federal courts convicted more than 20,000 government officials and private citizens involved in public corruption. The total convicted in the whole country is higher, since these numbers do not include convictions by states.

Changing Perceptions

By focusing on global governance standards that often reflect particular political, philosophical and ideological concepts of power, the traditional literature on governance has so far yielded few results. It has failed to offer actionable policies that poor countries could implement to foster inclusive growth in a pragmatic and incentives-compatible way.

In fact, good governance has been an elusive quest. Since the United Nations Commission on Human Rights identified transparency, responsibility, accountability, participation and responsiveness to the needs of the people as key attributes of good governance, the fight against corruption has become the most revealing and the most widely discussed aspect of governance. The academic pendulum on the subject has shifted from praising the increased economic efficiency that follows from corruption to stressing its many economic, sociopolitical and even moral costs.

Initial theoretical work on corruption underlined its positive role in development. Renowned scholars such as Nathaniel Leff (then at Harvard) and Samuel P. Huntington argued that corruption may make businesses more efficient by allowing them to circumvent bureaucratic procedures.

A second strand of the literature has attempted to invalidate these previous analyses. Gunnar Myrdal, the Nobel Prize-winning social scientist, argued that bribes may actually allow civil servants to reduce the speed with which they process business transactions, or that bureaucratic procedures should be seen not only as causes of rent-seeking activities but as their consequence.

Others have argued that, even taking at face value the suggestion that the most able economic agents in corrupt societies generate efficient allocation of resources through their actions, such talent allocation cannot be economically efficient. For instance, Susan Rose-Ackerman of Yale Law School observed that, once corruption is entrenched, it becomes so pervasive that it cannot be limited to areas in which it might be economically “desirable.”

A third and most recent strand of research has focused on the negative effects of corruption (and bad governance more generally) on economic growth. Kevin M. Murphy, Andrei Schleifer and Robert W. Vishny suggested that increasing returns on unproductive rent-seeking may eventually crowd out productive investment. Paolo Mauro offered empirical evidence that the prevalence of perceived corruption may negatively affect economic performance. Such problems are said to be even worse in countries rich in natural resources — especially those in the developing world — where opportunities for rent-seeking activities are typically very high.

Despite the insights from all these various waves of research, the problems of corruption and governance and their implications for economic development remain unresolved. Empirical demonstrations of the impact of governance on economic growth are often based on subjective perception indices, the limitations of which are well known. Policymakers in developing countries still have few actionable prescriptions for how to design policies to achieve their economic and governance goals. Moreover, a traditional recommendation for improving governance often involves curbing the power of political leaders — some of whom are not democratically selected. For low-income countries, a potentially more fruitful approach to tackling the problem would be to examine the possible determinants of good governance and to infer from these determinants which policies could limit opportunities for rent-seeking that in ways are compatible with political leaders’ personal goals.

The Cost of the Good Governance Rhetoric

The obsession that low-income countries must have the same political institutions as high-income countries is perhaps well explained through the story of a fake corruption scandal surrounding the visit of the president of Congo to New York in September 2005. His official purpose was to give a 15-minute speech to the UN General Assembly. But questions arose when the hotel bill for the president and the 56 people in his entourage was leaked to the press. President Denis Sassou-Nguesso had spent $295,000 for an eight-night stay at the Palace Hotel on New York’s Madison Avenue, including some $81,000 for his own suite.

The suite featured a master bedroom with a king bed, two additional bedrooms and six bathrooms. It also had its own private elevator. Media reports noted that the suite had a whirlpool bathtub and a 50-inch plasma television, and that room service charges on September 18 alone came to $3,500.

The hotel did not itemize the charges, leaving reporters to speculate about the room service menu, which included Dungeness crab, truffle crumbles, Scottish langoustines, pan-seared foie gras and braised snails in chicken mousse. They also made a big deal of the Congolese mission at the UN paying only a $51,000 deposit by check to secure the rooms. Presidential aides pulled out wads of $100 notes to settle the bill.

All the more stunning to reporters was that President Sassou-Nguesso was at the time the chairman of the African Union, representing the continent’s 53 countries, and also negotiating with the World Bank and the International Monetary Fund for the cancellation of a large fraction of Congo’s debt on the grounds that the country could not afford to repay it. His government was also talking to the Paris and London Clubs, informal groups of official and private creditors respectively, whose role is to coordinate sustainable solutions to debtor countries’ payment difficulties.

The news was received with shock and anger, especially by people in Congo, who probably would have liked their tax money spent on other priorities. Leaders of non-governmental organizations and anticorruption movements wrote letters to the World Bank’s president urging him to oppose any debt relief operation for Congo until the country’s leaders could demonstrate better public finance management skills. Global Witness, a well-known anticorruption group, issued a report claiming that Congo’s oil wealth has “for too long been managed for the private profit of the elite rather than for the benefit of its entire population.”

Not surprisingly, Paul Wolfowitz, then president of the World Bank, was more than inclined to bow to the pressure. It took a forceful response from the office of the executive director for francophone Africa on the board of directors of the World Bank to refocus the debate on Congo’s debt relief on the real issues at hand.

Let’s step back and look at the situation in the context of normal diplomatic practices. Why the outrage about a hotel bill of a few hundred thousand dollars for a large presidential delegation on an official visit to the United Nations? After all, hotel suites are expensive in New York in September, especially in the small number of luxury hotels where foreign heads of states are forced to reside for security reasons when they attend the annual UN meetings. Would those who cried foul about the hotel bill have preferred that the president of Congo and his entourage settle in a two-star hotel somewhere in New Jersey or Connecticut while attending the summit?

Is Incompetence Worse Than Corruption?

There are certainly many important economic and even governance issues to be discussed about Congo and other low-income countries. But the focus on the hotel bill obscured the real questions of whether the public policies implemented by his government were sound enough to bring strong economic growth and prosperity to his people. While the hotel bill might have been high, the only reason it was disclosed to the press was that some of Congo’s creditors had filed lawsuits over business debt repayment.

These were all “vulture” investment funds that make profits by buying up poor countries’ debt at discount prices. Using a judgment from two British High Court judges that found Congolese officials to be “dishonest” about their country’s debt, the fund managers had subpoenaed President Sassou-Nguesso’s hotel bill and leaked it to the media. Yet few newspapers that reported the sensationalist tale devoted time and resources to investigate the vulture investment funds — what they are, how they function and how poor countries around the world should deal with them.

Would the media have shown the same interest in the story if the hotel bill was run up by a leader from a country with a better reputation? Would these questions have arisen if the president of an industrialized country had spent the same amount of money for a stay in the city? Was the problem merely Congo’s intolerable poverty level? Or were the attacks against Sassou-Nguesso motivated by ignorance, class prejudice, racial prejudice and so on?

Perhaps because he had read news reports and briefing memos on issues such as the Sassou-Nguesso hotel bill story, Barack Obama used his first official trip to Africa (Ghana in July 2009) to speak out against corrupt leaders:

No country is going to create wealth if its leaders exploit the economy to enrich themselves or if police can be bought off by drug traffickers. No business wants to invest in a place where the government skims 20 percent off the top. … No person wants to live in a society where the rule of law gives way to the rule of brutality and bribery. ... And now is the time for that style of governance to end.

Those words were met with polite applause. But many African leaders and intellectuals objected to the paternalistic tone and the perceived double standard that underlined Obama’s public ethics lecture. Festus G. Mogae, the former president of the Republic of Botswana, observed that Obama’s critique of African corruption on his first official visit there seemed quite selective:

[Obama] has been to the Middle East, to Turkey, to Russia, to Europe, to Britain — Britain where Parliamentarians have been doing their own thing — [to] Germany where Siemens has been indicted for corruption, [to] Russia and the Middle East, [places] which are not known for their anti-corruption probity. … So, while it is right and proper that the president should have raised the issue of corruption, the fact that he only raised it when he got to Africa has the effect of perpetuating the misconception that corruption exists only in Africa.

The story of President Sassou-Nguesso’s hotel bill illustrates the confusion and fantasies that have too often plagued public policy when the good governance obsession leads to distracting public discourse and focus on the wrong development objectives. Such stories sideline the much bigger economic issues of public investment priorities, flawed debt management strategies and economic policy mistakes throughout the decades that are much costlier to Congo.

Similar stories can be told about many other countries. In the neighboring Democratic Republic of Congo (DRC), the public debate about corruption and good governance was dominated in 2013 by the story of 15 government officials who pocketed $52 million in mining fees in 2012. Again, that was a valid issue. But the public debate never tackled the much bigger problem: the DRC receives less than 5 percent of revenues generated by mining firms operating in the country, while the ratio is as high as 60-80 percent in the Persian Gulf countries and African countries such as Algeria. Honest incompetence and bad economic strategies are neglected, despite their potentially serious consequences on productivity and growth.

Likewise, the African Union has devoted many resources to promoting the findings of its Report on Corruption released in 2012, which indicates that an estimated $148 billion annually is lost to corruption. Clearly, such waste deserves publicity and reflection. But the sum should be put in the context of the much larger sums of money wasted on unproductive public expenditures.

An Incentives-Compatible Policy Framework for Governance

Most studies on the determinants of good governance go back to arguments similar to those made by either Gary Becker or Anne Krueger. Becker analyzed corruption as a purely illegal activity and suggested that criminal offenses must be viewed as “economic activities” with external effects, and with punishment conceived as a form of taxation. From that general framework he conjectured that the probability of committing a crime depends essentially on the penalty imposed and on the probability of being caught. Furthermore, the penalty’s deterrent value depends on the authorities’ willingness and capacity to enforce laws and on people’s acceptance of the country’s institutions. This implies that effective corruption enforcement rules, and good governance in general, can take place only in countries with political stability and transparent rules.

In Becker’s insightful analysis, corrupt agents expend resources to steal, and society, the victim, experiences negative external effects. He suggests that prohibition rules be combined with fines or other punishment to internalize the negative externality.

Unfortunately, this kind of after-the-fact remedy to corruption may arrive too late. And it may be ineffective in countries where the externality-generating activities (that is, corruption) are not easy to identify owing to prevailing social norms and practices, or may be costly to curb. In almost all poor countries, the costs of running a well-functioning national judicial system are often far beyond what the public sector can afford. The problem is compounded in many African countries where corruption is embedded in societal, economic and power relations, and virtually all state institutions, including the judicial system, are caught in the low-equilibrium dynamics of what Richard Joseph called “prebendal politics.”

But corruption isn’t just something that happens to poor countries. If one looks at corruption in historical perspective, it is clear that today’s high-income countries went through the same — or even worse — bad governance episodes now observed in sub-Saharan Africa. In the fascinating book Corruption and Reform, Harvard’s Edward Glaeser and Claudia Goldin analyze various patterns of bad governance in the history of the world’s greatest democracy. The results are disconcerting.

Conventional histories of 19th- and early 20th-century America portray its corrupt elements as similar, and at times equal, to those found in many of today’s modern transition economies and developing regions. Nineteenth-century American urban governments vastly overpaid for basic services, such as street cleaning and construction, in exchange for kickbacks garnered by elected officials. Governments gave away public services for nominal official fees and healthy bribes.

These elements provide a crucial clue to the problems of corruption and governance: they are endogenous to the level of economic development. In other words, low-income countries are by definition places where (perceived) corruption is a problem, while their governance indicators improve with their economic performance. It is unrealistic to expect the Democratic Republic of Congo, a country with less than $500 income per capita, to build governance institutions that are perceived as effective as those of Norway, where per capita income exceeds $70,000.

What is crucially needed, then, to fight corruption and improve governance in a low-income country is a development strategy that offers few opportunities for state capture and rent-seeking activities. If a government adopts a comparative—advantage strategy, firms in the priority sectors will be viable in an open, competitive market, and the government will not need to protect or subsidize them.

What does this leave us? Good political governance should be an important public policy goal and be set freely by the all countries’ people and leaders. It should not, however, be seen as a precondition for good economic performance. Sustained economic growth, employment creation and poverty reduction can be achieved even in very poor governance environments. Moreover, good political governance is always an unfinished process.

Good economic governance is a noble goal, and its general principles can be widely shared across nations and cultures. But operationalizing it is likely to vary widely across place and time. To succeed, a policy agenda is needed that focuses on using limited state resources strategically and wisely. Focus areas should include setting economic policy to ensure that only activities that are consistent with comparative advantage are encouraged; ensuring that government at all levels has the tools, incentives and discipline to facilitate public-private partnerships in the development of competitive industries; and setting rules of the game that are enduring and effective.

Following the policy-oriented approach suggested by Anne Krueger, the empirical literature has identified seven factors as the main causes of corruption:

• Trade restrictions, which make the necessary import licenses very valuable and encourage importers to consider bribing the officials who control their issuance.

• Government subsidies that are appropriated by firms for which they are not intended.

• Price controls whose purpose is to lower the prices of some goods below market value (usually for social or political reasons) but create incentives for individuals or groups to bribe officials to maintain the flow of such goods or to acquire an unfair share at the below-market price.

• Multiple exchange rate practices and foreign exchange allocation schemes. Differentials among these rates often lead to attempts to obtain the most advantageous rate, although that rate might not apply to the intended use of the exchange. Multiple exchange rate systems are often associated with anticompetitive banking systems in which a particular bank with strong political ties makes large profits by arbitraging between markets.

• Low wages in the civil service relative to wages in the private sector, which often lead civil servants to use their positions to collect bribes as a way of making ends meet — particularly when the expected cost of being caught is low.

• Natural resource endowments.

• Sociological factors such as ethnolinguistic fractionalization.

Given that virtually all governments in the world — including those in successful democratic countries — regularly intervene in their economies, the important question is which particular policy circumstances provide the best incentives for good governance? Lin’s book New Structural Economics attempts an answer. It suggests the gradual lifting of trade restrictions, price controls and multiple exchange rates, recognizing that such interventions may be temporarily needed to protect firms in sectors that lack a comparative advantage. It advocates carefully targeted incentives (of limited amount and time), allocated in a transparent manner to compensate for the externality generated by pioneer firms (even in industries that are consistent with comparative advantage).

Such a framework ensures that corruption opportunities are minimized. It favors government intervention only in industries where firms are viable in open, competitive markets and whose survival do not depend on large subsidies, or direct resource allocations through measures such as high tariffs, quota restrictions or subsidized credit. In the absence of large rents embedded in public policies, there will not be distortions that become the easy targets of political capture. The likelihood of the pervasive governance problems that are observed in many low—income countries can be much reduced if governments facilitate the development of new industries that are consistent with the country’s changing comparative advantage, determined by its changing endowment structure.

The goals of most political leaders everywhere are typically to stay in power as long as possible and to leave a positive legacy. Most leaders understand that promoting economic prosperity is the best way to achieve these goals. Development policy based on new structural economics, which advises governments to facilitate the entry of private firms into sectors with comparative advantages, can reduce corruption and bring growth. Good governance will be the result of such a strategy, because there is no need to create rents that subsidize and protect firms in the priority sectors. Therefore, it is an incentives-compatible way for political leaders in developing countries, including those in poor countries, to address challenging governance issues.

“Underdeveloped” Finance: The Illusions of Mimicry

On the list of the most recurrent obstacles to growth and poverty reduction, the next culprit often singled out in the literature is limited access to capital. Credit appears to be a major bottleneck for business creation and development. Moreover, bankers and financiers are universally considered villains whose greed and shortsightedness are such that entrepreneurs cannot expect their support in the drive for value creation.

In her compilation of jokes about them, Anna White (the City Diary editor of The Telegraph) recounts the widely shared belief that “bankers are people that help you with problems you would not have had without them.” She also tells the story of a man who visits his bank manager and asks, “How do I start a small business?” The manager replies, “Start a large one and wait six months.”

It is not surprising, then, that the weak financial systems that are so prevalent in low-income countries are perceived as impediments to growth and poverty reduction. Following the pioneering work of Raymond Goldsmith, Ronald McKinnon and Edward Shaw, a rich research literature has advanced the view that the amount of credit the financial sector can intermediate is an important determinant of economic performance.

Here again, the theoretical case seems quite strong. Economic prosperity, the result of improvements in physical and human capital and productivity, depends on the efficient use of productive assets — and on including large portions of the population in that process. Effective financial intermediation is therefore essential, as agents with net savings (whether domestic or foreign) should be encouraged to provide funding at optimal cost to support private investment. Both savers and investors face risk and uncertainty, and the financial system can help them mitigate it — or capitalize on it. By the same token, savers are generally unable to select the investment projects that best match their personal risk tolerances, and without pooling their money they cannot take advantage of increasing returns to scale in investments.

Ali Demirgüç-Kunt and Leora Klapper sum up the intellectual consensus on financial systems’ capacity to reduce poverty:

Inclusive financial systems — allowing broad access to financial services without price or non-price barriers to their use — are especially likely to benefit poor people and other disadvantaged groups. Without inclusive financial systems, poor people must rely on their own limited savings to invest in their education or become entrepreneurs — and small enterprises must rely on their limited earnings to pursue promising growth opportunities.

But empirical research tends to show that the relationship between financial development, economic growth and poverty reduction depends on many other factors, such as the country’s development level, the financial structure and existing regulations. Stephen Cecchetti and Enisse Kharroubi investigate one key question:

Is it true [that financial development is good for economic growth] regardless of the size and growth rate of the financial system? Or, like a person who eats too much, does a bloated financial system become a drag on the rest of the economy?

From a sample of developed and emerging market economies, they show that financial development promotes growth only up to a point, after which it actually reduces growth. Strikingly, they also show that a fast-growing financial sector is detrimental to aggregate productivity growth in advanced economies.

These findings raise a host of important questions for policymakers ranging from the criteria under which financial institutions are established and allowed to expand, to the instruments they can use to mobilize savings. Financial intermediation, everyone agrees, creates strong externalities that can be either positive (such as provision of information and liquidity) or negative (such as excessive risk-taking and systemic crises).

The Quest for Appropriate Financial Institutions

There is a vast literature analyzing the relative advantages of various banking structures. But there is no consensus on the strengths and weaknesses of alternatives in promoting economic growth. Nor is there consensus on the strengths and weaknesses of different financial regimes in different country settings. The reasons for these gaps are the neglect of the specific characteristics of the real economy at each level of development and the corresponding needs in terms of financial intermediation.

Research shows that financial markets tend to outpace banking activity as income per capita rises. The literature has focused on the causal relationship between financial structure and economic growth — that is, whether a market-based or bank-based financial structure is more conducive to growth. Banking partisans argue financial markets provide much weaker incentives for agents to collect information relevant to investors’ prospects before the fact or to monitor borrowers (or stock issuers) after the fact. Thus securities markets are at a disadvantage in alleviating the asymmetry of information between providers and users of capital. It follows that a bank-based structure should perform better in allocating resources and promoting economic development.

Not surprisingly, those who favor market-based financial systems focus on the problems created by powerful banks. Banks may gain huge influence over firms, undermining firm productivity and economic growth. In addition, banks tend to be more cautious by nature, and so bank-based systems may stymie financial innovation and impede economic growth. Furthermore, financial markets are often seen as providing richer and more flexible tools for risk management — derivatives and the like — while banks can provide only basic risk-management services.

The technical literature on banking tends to focus on whether competitive or monopolistic banks are better for economic growth. Some authors suggest that monopolistic banks tend to extract too much rent from borrowers (by charging above-market interest, etc.), undermining investment incentives. By the same token, monopolistic banks tends to pay lower rates of interest to depositors, and thus reduce the amount of capital they intermediate, which in turn slows economic growth.

But others argue that monopolistic banks have more incentive to collect information, screen and monitor borrowers, and form long-term relationships with borrowers. Such borrower-lender relationships are especially valuable to startups. In a competitive banking sector, by contrast, borrowers can more easily shift between lenders, so banks may have less incentive to forge such long-term borrower-lender relationships. As a consequence, market concentration in banking may give the most productive projects a better chance to get financed.

Empirical studies on this topic are far from conclusive. Some show a positive relationship between banking concentration and stability. Others find that new firms grow faster in economies with a more concentrated banking sector, while old firms benefit from a more competitive banking structure.

Despite their diverging conclusions, these two schools of thought share a similar — and flawed — research perspective. They typically start from an examination of various insti­tutional arrangements and then discuss the possible impact of financial structure on economic development. Yet the impact of financial systems on growth may not be appropriately determined if the analysis is isolated from the examination of how the financial structure itself is determined.

While the research on the impact of banking structure on market concentration is quite substantial, the topic of how the size distribution of banks affects growth has been neglected. This flies in the face of the well-established fact that small businesses, the dominant form of business operations in developing countries, usually have difficulty obtaining loans from big banks — implying that the distribution of banks by size does affect allocative efficiency.

A few studies have looked at the issue from a very different perspective, examining the mechanisms that determine financial structure. Raghuram Rajan and Luigi Zingales, for instance, apply interest group theory to explain differences in financial structure in countries at similar stages of development. Others have emphasized the legal system’s importance in shaping financial structure, arguing that effectively implementing the law is more critical to the operation of financial markets than of banks. Thus a bank-based system will have advantages in countries with weak legal systems.

This logic does not, however, explain why financial structure is usually different in countries with similar legal origins but at different development stages. Or why the financial structure in a given country changes as the country’s economy develops. In fact, any effective theory should take into consideration the financial structure’s endogeneity when analyzing the relationship between financial structure and economic development.

To sum up, much of the literature adopts a supply-side approach. It starts from an examination of the characteristics of various financial arrangements and then discusses the likely effects of different financial systems on economic growth. We think it is important to pursue a radically different demand-side approach — one that starts from the analysis of the real economy’s characteristics and the real economy’s demand for financial services. A financial structure’s effectiveness should be measured by one important criterion: whether it can best mobilize and allocate financial resources to serve the real economy’s needs.

Redefining Optimal Financial Structure

Empirical research shows that there is virtually no country — even among industrialized ones — where securities markets actually contribute a large part of corporate financing. Indeed, Colin Mayer studied eight developed countries and concluded that the average net contribution of their securities markets was close to zero. This does not necessarily imply that equity markets do not perform an important function, Mayer writes. “They may promote allocative efficiency by providing prices that guide the allocation of resources … through reallocating existing resources via, for example, the takeover process.”

Two dimensions of financial structure critically affect financial systems’ efficiency in economic development: first, the relative importance of banks and financial markets, and second, the distribution of banks of different sizes. Banks are the predominant type of financial intermediary in low-income countries. Their mechanisms for mobilizing savings, allocating capital and diversifying risks are very different from those of financial markets. Therefore, the relative importance of banks and markets constitutes the most important dimension.

Among banks, there is an obvious distinction between the way in which big banks do business and how small banks operate. This has implications for access to services, especially lending services, for different size firms. Banks were long regarded as central to promoting growth and poverty reduction. But in the face of widespread corruption and bank failures in the 1970s and 1980s, there was disillusionment with their role, most notably in developing countries. As a result, many influential development institutions, such as the World Bank, shifted their policy advice and advocated the use of both security markets and banks in promoting growth. In fact, a central feature of the economic reform implemented in the Washington Consensus framework was the dismantling of the traditional development finance model (based on bank-based systems, directed credit, public development banks, closed capital accounts, capped interest rates and active monetary intervention) that had been established in developing countries in the postwar era.

Small banks with very limited assets cannot afford to make large loans; they would have to bear much higher risk resulting from concentrated investments. Large banks are able to make larger loans and achieve better risk diversification. Since the transaction cost for making a loan is, at least to some degree, independent of loan size, large banks understandably prefer making loans to large firms rather than small ones. Large banks tend to focus on large businesses, while small banks specialize in lending to small businesses. Thus the distribution of big banks and their smaller counterparts can have a substantial effect on the banking sector’s performance.

The new received wisdom aims to reflect the imperatives of financial development. It has been influenced by financial market liberalization that is unfolding in the advanced economies, which have moved away from national bank—based systems toward open capital markets — at least until the 2008 Great Recession when the pendulum switched to the other extreme. Conservative governments in the United States and Europe abruptly changed gears and adopted new laws and regulations to rein in financial markets.

The financial-sector reforms implemented in developing countries around the world over the past decades were expected to raise savings and investment levels, reduce macroeconomic instability, increase the rate of growth and create employment. These objectives have generally not been achieved. Instead, the years since the mid-1990s have been marred by several financial crises and a decline in funding for productive enterprises — especially small and medium—size enterprises.

Recent analytic work by Justin Yifu Lin, Xifang Sun and Ye Jiang shows that each institutional arrangement in a financial system has both advantages and disadvantages. Equity markets become more active relative to banks as a country becomes richer; small businesses have no access to equity markets and generally have less access to large banks’ loan facilities.

The factor endowment in an economy at each stage of its development determines the optimal industrial structure in the real sector, which in turn constitutes the main determinant of the size distribution and risk features of viable enterprises, with implications for the appropriate institutional arrangement for financial services. Therefore, there is an endogenously determined optimal financial structure for the economy at each development stage.

While poor regulation and supervision may cause financial crises, a serious mismatch between the financial structure and industrial structure will reduce efficiency in mobilizing and allocating financial resources. In developed countries where large, capital-intensive firms and high-tech firms lead, financial systems dominated by securities markets and big banks will be more efficient in allocating financial resources and promoting economic growth. In developing countries, where small and less risky labor-intensive firms are the main engines of growth, the optimal financial structure will be characterized by the dominance of banks, especially small local banks. The optimal financial structure for any country changes as the economy develops.

The major policy challenge is selecting the appropriate framework for developing sustainable and effective financial institutions. In this regard, governments have an important role to play. Both equity markets and banks require regulation and supervision to reduce the occurrence and severity of financial crises. Although a country’s endowments and the resulting optimal industrial structure are the most fundamental force shaping its financial structure, government policy will also affect the evolution of the financial system. In fact, strategies promoted by governments are among the most important factors leading industrial structures and financial structures to deviate from optimality in terms of growth potential.

Policymakers in developing countries should be mindful of a particularly costly type of hidden distortion that follows from replicating the financial systems in developed countries without fully considering the real economy’s demand for financial services. Just like perception-based indicators of good governance, financial development is a function of a country’s economic development level — not a prerequisite to performance.