Chapter

2 Key Concepts, Benefits, and Risks

Author(s):
Jörg Decressin, Wim Fonteyne, and Hamid Faruqee
Published Date:
September 2007
Share
  • ShareShare
Show Summary Details
Author(s)

This chapter provides a bird’s-eye view of financial integration from a conceptual standpoint and also outlines its potential risks and benefits. At a very broad level, the integration of financial markets is a complex and multifaceted process, involving the behavior and interaction of agents, institutions, and markets operating in the context of different national policies and legal and prudential frameworks. In Europe, a layer of complexity is added at the supranational level through EU and euro area institutions—including the European Parliament, the European Commission, and the European Central Bank—and through policy initiatives—in particular, EU treaties and directives. To help make sense of all this, the conceptual framework developed here offers a lens through which to view many of the central policy challenges and issues facing the “European experiment,” upon which subsequent chapters elaborate. The following main questions are posed in this chapter: (1) What is financial integration? (2) What are the potential benefits? (3) What are the risks? Answers to these basic questions help define the key issues, the stakes, and the priorities for financial market policies.

What Is Financial Integration?

There is no single, universally agreed definition of financial integration. Structural definitions tend to emphasize the key attributes (institutional and policy-related) of integrated markets. These include features such as uniform rules, equal access, and equal treatment for market participants.1 Behavioral definitions typically involve the “law of one price.”2 Specifically, in an integrated system of financial markets, the price of fundamentally comparable assets or services should essentially equalize, reflecting an absence of arbitrage opportunities.3 Since the advent of Europe’s single currency, for example, euro area money-market yields have converged and, to a large extent, fixed-income markets have operated near “one price,” but credit markets—particularly for households—have not converged to the same degree (Figure 2.1; see also Chapter 3 and Baele and others, 2004).

Figure 2.1.Yield Convergence in EU Financial Markets, 1995–2005

(Cross-national standard deviations; in basis points)

Source: European Central Bank (ECB).

Note: The figure shows average dispersion of (unsecured) money market rates (overnight, one-month, and one-year lending rates), government bond yields at two-, five-, and ten-year maturity, and credit market rates based on various deposit and lending rates compiled by the ECB on a harmonized basis (and available only over a shorter sample); see www.ecb.int.

Neither type of definition, however, is wholly satisfactory. The behavioral viewpoint of financial integration fails to specify the key features, mechanisms, or ingredients needed to pursue integration, while the structural viewpoint fails to specify the criteria or metrics needed to assess whether integration is being achieved in practice. And neither accounts for the evolutionary dynamics associated with the changing degree of integration that characterizes ongoing developments in global and European financial markets.

In moving toward a more encompassing perspective, this book adopts the following broad definition: financial integration can be viewed as a multidimensional process, in which a system of financial markets becomes more closely interrelated over time in terms of its (1) market organization and infrastructure, (2) rules and regulations, and (3) pricing, transactions, and market practices.

From this “process-based” perspective, the degree of financial integration may best be viewed as proceeding along a continuum—from complete segmentation (financial autarky) on one end, to complete integration (single market) on the other (Figure 2.2). Financial autarky refers to a perfect separation of markets, operating essentially in parallel universes, without any linkage or interrelation among them. At the other end of the spectrum, a single market refers to the complete integration of markets, reflecting a full interconnectedness that makes geography irrelevant. In other words, any artificial dividing lines (such as political borders) that are unrelated to the specific functions or aspects of markets themselves would, in effect, vanish. In between these extremes, distinct markets may share some intermediate degree of interdependence, with developments in one or more financial markets partially affecting others. The defining features that characterize different points along this spectrum are elaborated here, through discussion of the evolving benefits and risks that accompany the process of broadening and deepening financial integration.

Figure 2.2.Spectrum of Financial Integration

Benefits of European Financial integration

Like money, finance is a public good—it enlarges society’s opportunities to consume and produce.4 At its essence, the benefits derived from finance are tied to economic growth, wealth creation, and welfare. Integration and its proximate benefits, in turn, can be seen in terms of the efficiency they grant in the provision of finance. These potential payoffs flow through several channels. Principal among these are greater efficiency in the allocation of resources (particularly, intertemporally); the formation of larger, more liquid markets; better diversification and risk sharing; and financial development and innovation. To be more concrete, the direct gains from integration can be viewed as operating along three key dimensions, which broadly align with Tobin’s (1984) concepts of efficiency in financial markets:5 (1) market access and competition; (2) market scale and structure; and (3) market scope and completeness. These integration-related concepts, their key market mechanisms for generating efficiency gains, and their broader economic impact are elaborated here.

Market Access and Competition

At a fundamental level, financial integration is about market access. Barriers to entry compartmentalize markets by restricting competition and contestability. Between countries, a large variety of such barriers exists, ranging from structural factors such as linguistic and cultural differences to economic factors such as a high fixed costs and network externalities. Integration policies take aim at reducing or removing key barriers—mainly, “artificial” or policy-induced barriers that are mostly of a legal, regulatory, or institutional nature. Such barriers impede local competition, limit external opportunities, and restrain access to financial services, and their removal or reduction is necessary to achieve closer integration of financial markets. Greater market access, in turn, can deliver (static) efficiency gains.6 For an existing market structure, for example, expanding the roster of potential entrants and making the playing field more level is likely to unleash new competitive pressures (through contestability or direct com-petition).7 This prompts existing firms to further pursue revenue and cost efficiencies, trim margins, and adopt managerial and organizational best practices, or else risk being supplanted by more efficient competitors or new entrants.8 Viewed from the inside out, market access also grants local firms the opportunity to look to other markets for new business activities and clients in order to expand their operations or the range of financial products they offer (including by tapping potential economies of scale and scope). Access and competition thus open avenues to arbitrage differences in the price and quality of (more accessible) financial services.

Beyond better functioning markets, what does greater access and competition in financial markets imply for agents’ welfare? For one thing, more efficient financial markets and greater access to financial services are likely to lower the costs of capital and improve the availability of funding for many types of activities, including venture capital and other forms of finance for start-ups.9 This, in turn, can foster an environment of greater risk taking that spurs market innovation and technological progress and that could have salutary effects on productivity and growth.10Pagano (1993), for example, argues that growth is enhanced through three channels: (1) by raising the fraction of savings funneled to investment; (2) by improving resource allocation across investment projects, thus increasing capital’s marginal productivity; and (3) by influencing households’ saving rates. While the net effect on the level of saving is generally considered to be ambiguous, improved intermediation (at lower cost) between savers and investors should produce a better allocation of saving and investment in the economy. For savers, this increases the (net) benefits from a given pool of funds. On the investment side, improved allocative efficiency mobilizes or channels funds more effectively into high-value projects. This enhances capital formation and the economy’s productive capacity.

Market Scale and Structure

Deeper financial integration also entails changes to the size and structure of the market itself. Integrating fragmented markets creates a combined market that is larger, deeper, and more liquid. Pooling liquidity, in turn, furnishes new trading opportunities, reduces margins or bid-ask spreads between buyers and sellers, limits the volatility impact of large trades, and contributes to more efficient price formation.11 As discussed in the previous paragraph, a larger overall market also provides existing firms (with a competitive advantage) the opportunity to realize greater scale economies, through higher volumes and lower average costs. Financial restructuring and consolidation (mergers and acquisitions)—which affect the number of firms in the market—reinforce this effect, enabling consolidated firms to create new synergies, reduce redundancies, and adopt more efficient scales of operation.12 In terms of market infrastructures (such as trading platforms and exchanges), consolidation may also produce network externalities in addition to scale economies.13

What bearing do market size and depth have on economic welfare? For one, a larger market and scale efficiencies may spur financial innovation—that is, the creation of novel financial instruments—that may have been infeasible in smaller, more illiquid markets, and that might provide new benefits to market participants. Along with technological progress and competition, this can strengthen dynamic efficiency in the financial sector and contribute to overall economic growth.

Comparing productivity developments in Europe and the United States, Box 2.1 shows that these two economies have performed very differently with respect to dynamic efficiency gains over the past decade or longer, with implications for relative economic growth and living standards. Examining this “productivity gap” at the sectoral level, it is clear that Europe’s financial sector has been partly culpable.

Market Scope and Completeness

Finally, financial integration may help complete “missing markets.” Financial innovation and the introduction of new financial instruments, for example, can create new types of finance and promote better hedging of certain risks at lower cost than previously possible. A notable example is the introduction of financial derivatives or other sophisticated financial products that offer new opportunities to manage and assume certain financial risks. In that sense, financial integration contains an important aspect of financial development, expanding the available set of economic opportunities offered by present market boundaries.14 For markets that are less sophisticated and developed initially, integration thus provides potential development benefits stemming from both financial “catch-up” and a faster pace of innovation.15 Financial integration also promotes greater risk sharing through diversification and the increasing ability of integrating markets to look beyond geographical borders. This provides new risk-management possibilities. Pooling of risks through greater country or regional coverage, for example, enhances resiliency to local (idiosyncratic) shocks. More efficient portfolio diversification, in general, should improve risk-return trade-offs, allowing investors to preserve their portfolios’ returns while reducing their overall riskiness.16

What does a more complete financial market do for welfare? The ability to better manage and diversify risks ultimately allows agents to better smooth consumption in the face of economic disturbances, and thus raises average welfare. For example, under a process of real integration that fosters regional specialization, national or local production may become more susceptible to local shocks.17 Financial integration, however, helps limit the variation in local income and consumption through greater risk sharing, including through diversified asset holdings, market insurance, and access to lines of credit.18 Market-based risk sharing along these lines is especially relevant for Europe, given the absence or limitations of alternative cross-regional mechanisms, such as fiscal federalism and transfers.19 Without more market-based risk sharing, risk-averse households would self-insure (through higher precautionary balances), which is less effective and more costly. Empirically, the evidence overwhelmingly suggests that risks are not fully shared internationally and that the potential gains may be very large.20Box 2.2 explores some coarse macroeconomic measures of risk sharing and financial integration in the European Union, including the sensitivity of countries’ private consumption to local shocks to national output.

Box 2.1.Explaining the United States-Europe Productivity Gap1

Productivity growth in the United States has outpaced that in the euro area since the mid-1990s, and the gap widened during 2001–04 (first figure). Lackluster productivity helps explain why—despite better job creation and improving labor utilization—per capita incomes in Europe have stagnated (if not slipped) relative to U.S. levels over the past decade (Estevão, 2004).

In terms of sectoral composition, the gap is mostly due to production sectors that are neither producers nor intensive users of information and communication technology (ICT). More widespread use of ICT in the United States, meanwhile, has resulted in strong productivity advances in key market services (van Ark and Inklaar, 2005).

TFP Growth in the United States and Euro Area

(In percent)

Note: Averages of Hedrick-Prescott-filtered data; TFP = total factor productivity.

GDP per Hour

(1990 PPP U.S. dollars)

Source: IMF staff calculations based on 60-Industry Database of the Growing Growth and Development Centre (www.ggdc.net).

Note: PPP = purchasing power parity.

Structural rigidities in Europe’s services sector, including financial services, are the main culprits for its sluggish productivity growth, as productivity in industry has kept pace. Barriers to competition and integration impede greater price flexibility and improved efficiency, including the uptake of new technology.

Data show that the financial sector accounts for almost half of the “growth gap” in aggregate labor productivity over the past decade, alongside wholesale and retail trade (second figure). Potential gains from promoting competition and further integration in the financial sector could thus be substantial. For example, estimates suggest that integrated clearing and settlement systems could reduce transaction costs by as much as 18 percent and increase GDP by around 0.6 percent (EC, 2006).

1 This box was authored by Emil Stavrev.

Figure 2.3 summarizes the main benefits of financial integration and the key mechanisms through which market access, scale, and scope potentially generate growth and welfare benefits.

Figure 2.3.Financial Integration and Its Benefits

Risks of European Financial Integration

While its benefits are numerous and far-reaching, financial integration is not without risks. Considering the various risks associated with any financial market, integrating markets accentuate those related to the transmission of shocks.21 With closer financial ties, the chances clearly increase that developments—including untoward events—in one financial entity or market will affect those in another. In this context, two related, but distinct, principal risks emerge: contagion and fundamental spillovers. This section describes the nature of these risks, their main transmission channels and components, and how they may evolve under a process of greater financial integration in Europe.

Viewing integration again along a continuum from autarky to a single market, Europe clearly lies somewhere in the middle, where both contagion risk and fundamental spillovers are present. Financial contagion may occur even with very fragmented markets (that is, at low levels of financial integration) because the broader transmission of shocks does not require fundamental ties. Fundamental spillovers, meanwhile, seem to require a higher threshold of integration. They tend to become more relevant as financial markets become increasingly interlinked and market interrelationships broaden and deepen (Figure 2.4). These issues become more apparent in the definition and elaboration of these financial risks and their sources and transmission mechanisms.

Figure 2.4.Financial Integration and Its Risks

Defining Key Financial Risks

The dividing line between financial contagion and fundamental spillovers is often blurred, both in theory and in practice. Just as for “financial integration,” there is no universally accepted definition of “contagion.”22Table 2.1 presents a representative list of alternative concepts of the term, which illustrates that the role of the “nonfundamental” dimension of contagion can differ significantly. Fundamental systemic risk, on the other hand, is usually defined in very broad terms, as risk that affects the entire financial market or system. In practice, all these risks may reflect interrelated aspects of the same phenomenon, broadly manifested by financial turmoil or instability.23 Moreover, they inevitably share similar traits—including the wider (possibly, cross-border) transmission of local disturbances—and operate through many of the same financial channels. Note too that these risks may also contain a global dimension given the influences and linkages between the world’s financial markets.24

Table 2.1Alternative Definitions of “Contagion”(In order of increasing specificity)
DefinitionCommentSource
Cross-country transmission of shocks or general cross-country spillover effects“Broad” definitionWorld Bank (n.d.)
Volatility of asset prices that spills over from the source country to others countriesPericoli and Sbracia (2003)
A particularly strong propagation of failures from one institution, market, or system to anotherde Brandt and Hartmann (2000)
Excess correlation—that is, correlation over and above what one would expect from economic fundamentalsBekaert and Harvey (2003)
When the extent and magnitude to which a shock is transmitted internationally exceeds expectationsEdwards (2000)
Transmission of shocks to other countries, or cross-country correlation, beyond any fundamental link among the countries and beyond common shocks“Restrictive” definitionWorld Bank (n.d.)
When the transmission channel intensifies or changes, contingent on a crisis(Shift-) contagion definitionPericoli and Sbracia (2003)
A significant increase in the comovements of prices and quantities across markets, conditional on a crisis occurring in one marketPericoli and Sbracia (2003)
Cross-country correlations that increase during “crisis times” relative to correlations during “tranquil times”“Very restrictive” definitionWorld Bank (n.d.)

To be more concrete, this chapter employs the following working definitions of the key risks associated with financial integration.

  • Financial contagion refers to an adverse external or spillover effect transmitted across financial markets beyond what fundamental linkages or factors—including common shocks—would predict.25 Contagion effects themselves can occur on multiple levels, including through market sentiment or activity, for example through prices, volumes, or liquidity. But the key distinguishing feature is the second part of the definition, which conveys an important nonfundamental basis for contagion risk—that it exceeds what might be warranted based solely on underlying market fundamentals.26 Akin to catching the flu or a virus, contagion involves the transmission, through either direct or indirect contact, from one financial entity to another, where the latter was deemed to be otherwise “healthy.”
  • Fundamental spillovers refer to the wider spread of financial disturbances through market relationships, transactions, or exposures, reflecting fundamental linkages or interdependencies between various entities or markets. To the extent that these financial risks ultimately jeopardize the functioning of the entire system, they would constitute a systemic risk.27 Like a cancer that spreads, systemic risk may originate in a specific locality but, by spreading through a network of financial interrelationships, it ultimately threatens the entire financial system. This could have potentially grave consequences for the real economy.28 Common shocks can be an important source of systemic risk.

Box 2.2.EU Consumption-Output Correlations1

The sensitivity of consumption to output and idiosyncratic shocks constitutes a broad measure of macroeconomic risk sharing, expected to become lower the greater the degree of risk sharing an economy undertakes with other countries. This sensitivity can be examined through a panel specification that regresses countries’ relative growth in private consumption on their relative growth in GDP:

The coefficient related to consumption smoothing, b, is estimated to be 0.42 percent, pointing to less than full risk sharing in the European Union. The results are based on quarterly data for 15 EU countries from the second quarter of 1995 to the fourth quarter of 2005.2 The data provide no evidence of increased consumption smoothing, as the estimated value of b rises to 0.48 over the latter half of the observation period.

Another perspective comes from saving-investment correlations and the well-known “Feldstein-Horioka puzzle.” Financial integration should weaken the link between national saving and investment, as it enables EU member states to engage increasingly in intertemporal trade. Many authors dispute that a high level of correlation necessarily reflects weak integration, because shocks (for example, to productivity) may affect both investment and saving in the same direction. Nonetheless, changes in the correlation may be illustrative, if underlying shock processes have not materially changed. The basic regression is

Blanchard and Giavazzi (2002) show that the saving-investment correlation for the euro area—measured by the coefficient d—declined in the 1990s, when compared with the period 1975–90. They also report smaller reductions for 22 country members of the Organization for Economic Cooperation and Development (OECD) and the European Union. The table below applies their methodology to a more recent data sample with redefined subperiods. The results corroborate the findings of Blanchard and Giavazzi.

Estimated Feldstein-Horioka Coefficients, 1980–20031
PeriodOECD2EU3Euro Area4
1980–20030.620.560.66
1980–19920.720.760.83
1993–20030.500.450.51
Source: IMF staff calculations.

Results of regressions of investment on saving, both as a percentage of GDP, including country-specific dummies.

All OECD countries except the Republic of Korea, Luxembourg, Mexico, Turkey, Czech Republic, Hungary, Poland, and Slovak Republic. The final sample contains 22 countries.

EU data refer to EU-15 minus Luxembourg.

Euro area series excludes Luxembourg.

Source: IMF staff calculations.

Results of regressions of investment on saving, both as a percentage of GDP, including country-specific dummies.

All OECD countries except the Republic of Korea, Luxembourg, Mexico, Turkey, Czech Republic, Hungary, Poland, and Slovak Republic. The final sample contains 22 countries.

EU data refer to EU-15 minus Luxembourg.

Euro area series excludes Luxembourg.

1 This box was authored by Marcello Estevão and Chanpheng Dara.2 This estimate lies below the estimate in Adam and others (2002) using annual data. However, in their specification, common movements in GDP across the European Union were not included. Hence, sensitivity of national consumption to these aggregate EU-wide changes may also be reflected in their point estimate.
  • Transition risk is a third type of risk that can encompass elements of the other two but is reserved for the specific case of evolving risks arising from changes in the degree of integration, rather than those associated with a particular level of integration. In Europe, this risk is in the same direction. Nonetheless, changes in the correlation may be illustrative, if underlying shock processes have not materially changed. The basic regression is particularly germane for “catch-up” countries that experience a sharp fall in interest rates and pronounced credit booms during a phase of real and nominal convergence and increasing financial integration accompanying EU or euro area membership.29

Operationally, it is difficult to distinguish among these risks for the aforementioned reasons. Nevertheless, this broad classification is useful conceptually to help identify more completely possible sources of financial risk or mechanisms that generate risk, to track how these might evolve over time as the integration of European financial markets proceeds, and to identify the corresponding requisites for financial sector policies.30 The next section elaborates on the key transmission routes for these risks.

Transmission Channels for Financial Risks

The main channels for fundamental spillovers with possible cross-border dimensions include the following:

  • Financial consolidation, competition, and stability. The nexus between competition and stability in the banking system has been the subject of a long-standing policy debate.31 Some have suggested that less competition—accompanying consolidation and rising market concentration—improves financial stability. They argue that market power and higher profits act as “buffers” to help firms weather adverse market conditions, shun riskier ventures, reduce turnover, and avoid liquidity problems.32 Others, however, argue that consolidation might lead to
    • – the creation of financial firms “too large to fail, liquidate, or discipline” with a greater appetite for risk (moral hazard issues) and a greater likelihood of disorderly workouts;33
    • – a higher average size of bilateral exposures, and concentration among fewer firms, which may potentially cause greater fallout in the event of financial stress;34
    • – higher rates charged by monopolistic firms, which could push borrowers toward greater risk-taking and lead to adverse selection among them;35 and
    • – firms that look more alike, both in their geographical presence and their range of financial products, and as a result exhibit increased correlation in their responses to aggregate shocks, which could magnify the effects of such shocks.36

In Europe, technology, globalization, and deregulation have been powerful forces driving restructuring and consolidation in line with global trends.37 But, until recently, consolidation in the financial sector has occurred largely within rather than across national borders, more so than in other (nonfinancial) sectors.38 Padao-Schioppa (2000) argues, however, that this national consolidation is still consistent with the pressures emanating from greater financial integration across European markets.39

  • Financial conglomeration, complexity, and transparency. A parallel but distinct financial trend is conglomeration, where different types of financial institutions enter into each other’s territory (horizontal diversification). The major risk-reducing benefit stems from increased product and revenue diversification and potential economies of scope.40 Financial risks associated with conglomeration revolve around the following considerations:
    • – The emergence of large, and increasingly complex, financial institutions raises questions about market transparency and the ability to monitor these firms effectively. The Bank of Credit and Commerce International (BCCI), for example, managed to escape effective external oversight through a complex corporate structure spread out over numerous countries.41 The problems surrounding the Long-Term Capital Management (LTCM) crisis also extended beyond leverage and liquidity issues to the complexity and opacity of LTCM operations, which thwarted effective market discipline.42
    • – The flip side to economies of scope is that risk modeling and management might be harder for institutions venturing into a wider range of (noncore) activities, because the complexity of risks rises substantially and appropriately aggregating these risks may be difficult.43
    • – With growing organizational complexity, operational risks—related to aspects of the organization’s structure and operations, including effective internal monitoring and control—become more challenging, as highlighted by the collapse of Barings Bank in 1995.44 This and other cases illustrate clearly that insufficient checks and balances can allow individual employees scope to cause tremendous damage.

Conglomeration as a financial market trend has also figured prominently in Europe in recent years. Many banks, securities firms, and insurance companies have merged into financial conglomerates.45 As with consolidation, conglomeration trends have until recently largely occurred within the national domain in Europe.46

  • Cross-border ownership and related exposures. Integration increases possibilities for stronger balance sheet linkages and exposures through increased cross-border ownership stakes. As a result, changes in the market valuation of financial firms in a particular location could have wider (cross-border) repercussions. Moreover, ownership links often lead to concentration in other balance-sheet exposures, such as (intragroup) credit, as well as off-balance-sheet exposures, such as guarantees. These links constitute additional transmission channels for financial disturbances across national markets. Until recently, the relevance of these channels was limited, because financial consolidation and conglomeration initially occurred largely within and not across European borders. But recent trends and the pressure exerted by market forces indicate that this is changing, even though certain barriers remain in place.47
  • Cross-border lending and payments. Ownership issues aside, increased transactions between financial firms—particularly, banks—raise the possibility that distress in one institution could impinge upon others through disruptions in the payments system and credit—in particular interbank—markets.48 Distress could spread both through liquidity and credit channels. For example, from a liquidity perspective, difficulties at one individual bank may initiate a “domino effect” if that bank’s nonpayment or payment delays on its interbank claims jeopardize the ability of its creditor banks to meet their obligations to their creditors.49 In addition to this liquidity risk, losses on credit exposures—be it to other banks, nonfinancial firms, or households—constitute another possible cross-border transmission channel. Thus far, notwithstanding the advent of the euro and price convergence, interbank exposures remain disproportionately at the national level. About half of interbank transactions are cross-border (see Chapter 3), but these exposures tend to be more dispersed than intraborder exposures. Consequently, for now, risks from interbank exposures vis-à-vis banks in other European countries remain significantly below those from domestic interbank linkages.50 Other exposures also remain concentrated at the national level for most banks.51 But both the number of banks that engage in cross-border credit and the extent of their exposures are on the rise and some concentration of cross-border operations is occurring, for example, in London as a financial center and in the growth markets of Central and Eastern Europe.
  • (Fundamental) portfolio rebalancing. Increasing portfolio diversification (that is, declining “home bias”) increases the likelihood that the effects of portfolio rebalancing in one market will be transmitted to other financial markets. Local shocks, such as capital losses in one market, may induce “contingent selling” of other assets, for example, to raise collateral or liquidity, or to meet margin calls.52 Portfolio investment data suggest that diversification and cross-border holdings of European equities have increased among both EU and euro area investors.53

Financial Contagion

Beyond these fundamental spillovers, an additional source of risk and the transmission of financial disturbances is contagion. Unlike fundamental spillovers, financial contagion operates through nonfundamental channels that may become more important with market integration, including the following:

  • Incomplete information and herding. With imperfect information, price changes in one market are perceived as having implications for the value of assets in other markets, causing their prices to change as well (King and Wadhwani, 1990). This cross-market signaling effect between, say, European markets might intensify as the perceived relevance of one market’s order flows for other markets increases with integration. In the case of information flows, more highly integrated markets can be expected to disseminate information more rapidly to all market participants. This greater fluidity could also apply to misinformation or “rumors,” with implications for speculation and volatility.54
  • Self-fulfilling expectations and panics. Bank runs can occur when depositors not facing liquidity shocks decide to withdraw funds after observing others withdrawing their funds, in the fear that they will not be able to recover their deposits. These beliefs can then become self-fulfilling (see Diamond and Dybvig, 1983). With more integration in banking, liquidity concerns (warranted or unwarranted) could become more intermingled among various institutions.
  • (Nonfundamental) portfolio rebalancing. Asset reallocation may lead to “contagious selling.” An unfavorable market shock that reduces wealth, for example, may lead to an increase in risk aversion and a general pull-back from risky assets, thereby impinging on other markets.55

Fostering Integration and Monitoring Financial Risks

As Europe’s system of financial markets continues to evolve, so too will the risks. To the extent that capital markets and finance in Europe take on an increasingly cross-border dimension, integration will introduce risks that are not yet known and that are more likely to spread—whether through confidence, balance sheets, lending and payments, or other mechanisms. Moreover, the complexity of institutions with cross-national organizational structures may make them more opaque to outside monitors (from multiple countries).56 But the potential gains from integration in terms of efficiency and welfare are vast. And the right path is clear. Europe needs to seize the benefits that integration has to offer, as effectively as possible, while managing the attendant risks. Policymakers will need to stay a step ahead of this dynamic process, modulating or reforming the regulatory, supervisory, and crisis management frameworks as needed to promote market efficiency while safeguarding financial stability. The question before us, and the subject of this study, is how best to chart such a course for Europe.

2See, for example, Adam and others (2002), who discuss various integration measures and their relative merits, including not only price- but also quantity-based measures (such as cross-holdings and diversification) and legal and institutional measures (such as dispute resolution).
3In the case of equities this involves issues related to the comparability of underlying assets. Levy-Yeyati, Schmukler, and Van Horen (2006) control for this by examining the “cross-market premium” or price differential of cross-listed stocks on multiple exchanges as a measure of financial integration.
4But whereas “money”—as a medium of exchange—eliminates the need for “double coincidence of wants” and grants “finality of payment” in instantaneous trade and exchange, “finance” involves an intertemporal exchange between parties, trading liquidity for a future higher return. Thus, Schinasi (2006) notes that “finance involves uncertainty and risk about human trust,” but has the potential to create a superior store of value.
5Tobin distinguishes between several efficiency concepts for a system of financial markets, including technical efficiency and information arbitrage, functional efficiency, and full-insurance efficiency (à la Arrow-Debreu). See, for example, Tobin (1984) and Buiter (2003).
6Based on evidence from 47 countries and 1,200 banks, Levine (2003) finds that denying foreign entry raises interest margins. Casu and Girardone (2006) find that increased competition may have forced EU banks to become more efficient, but the reverse does not appear to hold—that is, more efficient EU banking systems are not necessarily more “competitive.” Instead, (cost-)efficient banks have acquired less-efficient rivals and concentration ratios have not declined between 1997 and 2003.
7See, for example, Claessens and Laeven (2004).
8Using bank-level data for 80 countries for 1988–95, Claessens, Demirgüç-Kunt, and Huizinga (2001) find that foreign entry reduces profitability and overhead expenses in domestic banking. See also Levine (2003); Claessens and Laeven (2004); and Demirgüç-Kunt, Laeven, and Levine (2004). Hasan, Malkamaki, and Schmiedel (2003) find that increased competition raises cost and revenue efficiency of stock exchanges.
9Looking at deposit contracts, Fecht and Martin (2005) argue that competition from other banks reduces rents and can be welfare improving, but competition from financial markets—that is, increased household access to nonbank markets—may reduce the risk sharing offered by banks through a substitution effect.
10London Economics (2002) examines (static) gains from financial integration, finding that it would durably lower the cost of capital by ½ percent and boost EU GDP by 1.1 percent in the long run. Using the Rajan-Zingales (1998) methodology, Giannetti and others (2002) find that EU financial integration—assumed to raise firms’ access to financial markets to U.S. levels—could increase European manufacturing growth by ¾ percentage point a year.
11See, for example, European Commission (2002) for a discussion in the context of the Investment Services Directive (ISD).
12Carletti, Hartmann, and Spagnolo (2003) find that bank mergers increase aggregate liquidity (due to scale economies) if interbank refinancing is more expensive than financing through deposits. There are studies, however, indicating that mergers of very large banks produce little or nothing in the way of scale economies and may even produce diseconomies. See Group of Ten (2001) and De Nicolò (2000) and references cited therein.
13Hasan and Schmiedel (2004) examine network benefits in the case of European equity markets. Schmiedel, Malkamäki, and Tarkka (2006) find potential scale economies in European securities depository and settlement systems.
14Shiller (1993, 2003) presents a visionary exposition on the scope for further financial innovation and risk management and, by implication, on the prevailing high degree of financial market incompleteness.
15Giannetti and others (2002) find wide dispersion in financial development in the European Union. Following Rajan and Zingales (1998), they find these financial development “gaps” matter for growth. For a recent survey on financial development and growth, see Levine (2004).
16These issues are taken up in Chapter 5.
17Kalemli-Ozcan, Sorensen, and Oved Yosha (2001, 2003) find that industrial specialization has intensified within the European Union owing to comparative advantages and that, as a consequence, GDP asymmetry has increased.
18Assuming that country-level GDP follows a random walk, Van Wincoop (1994) finds that the gain from full risk sharing would be equivalent to a permanent increase in consumption of about 2 to 3 percent. Cochrane (1991) and Mace (1991) note that if idiosyncratic risk is fully shared, then individual consumption should be affected only by aggregate fluctuations. They test this proposition using household data for the United States.
19Mélitz (2004) studies transfers from the U.S. federal government to explain state-level income smoothing. See, among others, Sala-i-Martín and Sachs (1992); Asdrubali, Sorensen, and Yosha (1996); Sorensen and Yosha (1998); Mélitz and Zumer (1999); and Kalemli-Ozcan, Sorensen, and Oved Yosha (2001, 2003, 2004).
21Typical financial risks include liquidity risk, credit risk, market risk, and operational risk. See Schinasi (2006) for a discussion.
23See Schinasi (2006) for a review of the alternative definitions of (and sources of risk to) financial stability.
24Baele and others (2004), for example, suggest that the “spillover intensity” or the sensitivity in European financial markets to global (U.S.) shocks has been rising—indicative of increasing global market integration.
25This economic definition essentially follows the World Bank’s “restrictive” definition (World Bank, n.d.).
26The related issue of whether contagion is “rational” or “irrational” is not without controversy. “Pure contagion,” often reserved for nonfundamental phenomena, includes disequilibria due to market imperfections (for example, incomplete information, noise traders, herding as in King and Wadhwani, 1990) or multiple equilibria (for example, sunspots or self-fulfilling crises as in Masson, 1999). But several studies have also put forth notions of “rational” contagion.
27Note that contagion, narrowly defined, is unlikely to constitute a systemic event solely on its own, since that would require a contagious, systemwide panic or loss of confidence not grounded in common shocks or fundamental factors. Nevertheless, contagion may trigger or exacerbate events that could ultimately disrupt the functioning of the financial system, and thus comprise an important contributing factor to systemic risk.
28Several studies require that risks pose a potential serious impact on the real economy in order to be classified as “systemic.” See, for example, De Nicolò and Kwast (2002) and Group of Ten (2001).
30De Bandt and Hartmann (2000) argue that while “…it might be futile to look for the single, ultimate definition of systemic risk, it may still be useful to give some general structure…in order to help avoiding piece-meal policy making.”
35See, for example, Boyd and De Nicolò (2005). Although market concentration ratios have generally been rising, the evidence is less clear that competitiveness and efficiency levels have been significantly affected. On the complex link between market concentration, competitiveness, and fragility in the banking system, see Casu and Girardone (2006); Beck, Demirgüç-Kunt, and Maksimovic (2003); and Demirgüç-Kunt, Laeven, and Levine (2004).
36On systemic risks attached to greater similarity and size, one indication of susceptibility to common shocks is given by the correlation in firms’ equity returns. Based on the comovements in the growth of bank share prices, however, there is no clear evidence of an increase either within or across European banking systems during the 1990s (Group of Ten, 2001). De Nicolò and Kwast (2001) find the opposite result for the United States. See also Chapter 6.
37See Berger and others (2000) for a survey. From 1997 to 2003, almost 2,200 EU credit institutions (around 23 percent of the 1997 total) ceased to exist, almost entirely as a result of mergers and acquisitions. In 2003 alone, the number of credit institutions declined by 4 percent in the EU-15. Moreover, concentration ratios in financial markets have generally risen: on average, the five largest institutions’ share of total bank assets amounted to 53 percent of the total in the EU-15 in 2003 (on an unconsolidated basis), up from 46 percent in 1997. See European Central Bank (2004e).
38Less than a quarter of mergers and acquisitions within the banking sector have been cross-border, a low figure compared with other sectors. See Cabral, Dierick, and Vesala (2002); European Commission (2002); Baele and others (2004); the European Commission’s Financial Integration Monitor (2004); and European Central Bank (2004a). Since 2004, though, there has been some notable merger and acquisitions activity. See Chapter 6.
39In a contrarian view, Berger and others (2000) and Berger, DeYoung, and Udell (2001) posit that the observed infrequency of cross-border mergers reflects the subpar performance of foreign-owned banks in developed markets, given structural barriers such as culture and language.
40A well-known finding though is a “diversity discount” in the market valuation of industrial conglomerates (around 8–15 percent in the United States). See Lang and Stulz (1994) and Berger and Ofek (1995). This may reflect internal resource misallocations (Rajan, Servaes, and Zingales, 2000). Elsas, Hackethal, and Holzhauser (2006), however, find that revenue diversification for banks increases profitability and market value, unlike the typical finding for industrial firms. In the case of Europe’s banks, branching out allows them to diversify their customer bases and increase customer loyalty by offering a fuller range of financial products, while improving the risk-return structure of their assets and liabilities. Moreover, a shift away from low-margin, core activities—such as taking deposits and making loans—may allow banks to partly shift their role as principal (bearing the risk) to that of agent (acting on behalf of the principal), depending on the services rendered. This could allow, for example, an increasing role for fee-based revenue streams that do not put bank capital at stake.
41See, for example, Herring (2004).
42See Group of Ten (2001). Greenspan (2005) emphasizes that LTCM’s large net positions in illiquid markets undermined the usefulness of collateral as a risk mitigant: when nondefaulting counterparties attempted to close their positions, rapid swings in prices amplified the losses to LTCM.
43Boyd, Graham, and Hewitt (1993) find that combining banking and insurance activities reduces earnings volatility, but Allen and Jagtiani (2000) do not find a similar benefit from combining interest-based banking and fee-based securities activities. Laeven and Levine (2005) find that financial conglomerates have market-to-book values that are roughly 10 percentage points lower than those of comparable but focused interest- or fee-based firms.
45In addition, many financial firms have sought to leverage their name recognition and customer franchise by broadening their product range beyond their traditional area of expertise. Banks have started to offer insurance products, for example, while traditional banking products (e.g., mortgages) are now being offered by some pension funds and insurance companies. See European Central Bank (2004a, 2004b).
46Bank-insurance or “bancassurance” mergers and acquisitions over the past decade have been conducted essentially within national borders, and after a spurt of activity at the end of the 1990s and early 2000–01, the pace has waned in recent years. See European Central Bank (2004a).
47Banking studies find that beyond structural factors (distance, language, information costs, etc.), government regulation is an important determinant of international merger decisions. See, for example, Buch and DeLong (2001).
48Banks internalize the fragilities associated with liquidity provision (maturity transformation) and hence are themselves prone to such fragilities (Diamond and Rajan, 2001; and Schinasi, 2006). Nowadays, however, banks have many more tools at their disposal to hedge these risks than in the past.
49Allen and Gale (2000b) and Rochet and Tirole (1996) examine the vulnerability of banks to a chain reaction or “domino effect” through interbank lending and deposits.
50In France and Germany, for example, domestic interbank loan volumes remain four and eight times larger than loans to banks in other euro area countries. Bank failures would therefore likely affect domestic counterparts in the first instance and these counterparts also risk being less diversified in their interbank exposures. The United Kingdom is a notable exception. Here (unconsolidated) cross-border interbank assets in other European countries amount to about 30 percent of the total assets of the domestic banking system, about the same magnitude as domestic interbank loans. Cross-border interbank assets held in non-European banks account for another 25 percent of total assets. See European Central Bank (2004e).
51Buch, Driscoll, and Ostergaard (2004) find European banks’ asset portfolios are domestically overweighted (or underdiversified from a cross-border perspective).
53See, for example, De Santis and Gerard (2006). Portfolio diversification and home bias issues are discussed at length in Chapter 5.
54Calvo and Mendoza (2000) suggest that the globalization—here, euro-ization—of securities markets exacerbates herding behavior (and volatility), due to the higher cost of collecting information.
55See Kyle and Xiong (2001). See also Kodres and Pritsker (2002), who examine cross-market rebalancing under information asymmetries, showing that “excessive” asset price movements across markets occur relative to a full-information setting.
56Another issue with cross-border consolidation arises when locally operating branches of foreign banks are systemically more important to the host country than to the home country. For example, the failure of a bank from a large country that has a branch in a small country may have larger systemic concerns in the latter, where the branch may be large in relation to the financial system and the economy. But according to the “home-country principle,” it is the responsibility of the supervisory authorities or central bank from the large country to decide whether or not to intervene. Therefore, financial stability of smaller countries may be vulnerable to the behavior of foreign banks (and the incentives and decisions of foreign regulators), and domestic authorities may also have limited powers in the event of a systemic situation. See Chapter 10.

    Other Resources Citing This Publication