Economic Vulnerability of Least Developed Countries in Light of the Current Economic Crisis

Table of Contents

1. Introduction

2. Least Developed Countries
2.1. Definition
2.2. Structural Features

3. The Economic Development from 2000 to 2007
3.1. Decoupling or Convergence?

4. The Economic Crisis of 2008
4.1. Anatomy of the Subprime Mortgage Crisis
4.2. Valuation Aspects and Asymmetric Information
4.2.1. Adverse Selection between Individual Actors
4.2.2. Moral Hazard between Individual Actors
4.2.3. The Role ofBanks
4.3. Impact and Effects on the World Economy
4.4. Impact on Developing Countries
4.4.1. Direct Financial Channels
44.1.1. Financial Openness
44.1.2. Financial Institutions and the Real Economy
4.4.2. Spillovers from the Global Recession
44.2.1. Foreign Direct Investment
44.2.2. Aid Flow Volatility
44.2.3. Commodity Trading and Commodity Price Volatility
4.4.3. Preliminary Conclusions

5. Economic Vulnerability
5.1. Definition and Use
5.1.1. EconomicApproach
5.1.2. Vulnerability in the Context of Political Science
5.2. Review of Existing Indices
5.2.1. Economic Characteristics of Small Size
5.2.2. Discussion
5.3. Construction of an Economic Vulnerability Index
5.3.1. CautionaryNotice
5.3.2. Methodology
5.3.3. Indicators of Structural Vulnerability
5.3.4. Exposure
5.34.1. Population Size
5.34.2. Location Index
5-3-5- Shock Index
5.3.5.1. Trade Shock Index
5-3.5.2. Financial Shock Index
5.3.6. Substitutability Index
5.3.6.1. Merchandise Export Concentration
5.3.6.2. Dependence on Strategic Imports
5.4. Dynamic Examination of Vulnerability Indicators
5.4.1. Financial Openness Index
5.4.2. Terms ofTrade
5.4.3. Dependence on Strategic Imports Index
5.4.4. Export Concentration Index
5.4.5. Preliminary Conclusions
5.5. Economic Vulnerability Index 2006
5.5.1. Limitations ofIndices

6. Conclusion

References

Annex A: List of Least Developed Countries

Annex B: Data Availability

Annex C: Exact Calculations of the Economic Vulnerability Indicators

Annex D: Complete List of Vulnerability Indicators used in the Composite Index

List of Tables and Figures

Figure l: Composition ofMerchandise Exports in LDCs

Figure 2: Indicators ofEconomic Development

Figure 3: Case-Shiller House Price Index

Figure 4: A-Rated Home Equity Loans ABS Fitting Errors

Figure 5: Aid Inflows as Percentage of Gross National Income

Figure 6: Commodity Price Volatility and LDCs Exports

Figure 7: Classification ofVulnerability

Table 1: Overview of EconomicVulnerability Indices

Figure 8: Composition ofthe EconomicVulnerability Index

Table 2: Correlation Matrix for the Financial Openness Index

Figure 9: Financial Openness Index

Table 3: Density Mass of Financial Openness Indicator up to a Critical Value

Figure 10: Terms of Trade Density Distribution

Table 4: Correlation Matrix for Terms ofTrade

Figure 11: Dependence on Strategic Imports Index

Table 5: Correlation Matrix for Dependence on Strategic Imports Index

Table 6: Density Mass of Dependence on Strategic Imports Indicator up to a Critical Value

Figure 12: Export Concentration Index

Table 7: Correlation Matrix for Export Concentration Index

Table 8: Correlation Matrix for DifferentValues of b

Table 9: EconomicVulnerability Index (Different Specifications)

Abstract

As the economic crisis of the late 2000s unfolds, the need for global political, regulatory and supervisory changes becomes apparent. This paper is intended to clarify the role that least developed countries play in this process. Different channels of vulnerability are identified and examined both in a dynamic and static framework, thereby contributing to an understanding of how and to which degree the world's poorest countries will be affected by the current global economic turmoil. It arrives at the conclusion that LDCs are likely to be negatively affected by numerous structural factors and identifies low and high vulnerability states by constructing a composite Economic Vulnerability Index.

l. Introduction

II n'y a de nouveau que ce qui est oublie.

— Marie-Jeanne Bertin (1747-1813)

In recent history, only one event resembles the uproar generated by the current financial market crisis and the consequential world wide economic distress: the Great Depression. In that sense, there is indeed nothing new except for what has been forgotten; well, almost nothing. Various notable economists, among them Nouriel Roubini, Paul Krugman and Ravi Batra, have been quick to draw comparisons between 1929 and today, implying that the world will experience its worst recession since the Great Depression. Although these statements may be correct, they fail to capture the extent to which the two events differ. The effects of the current crisis are likely to be more widespread and diffuse than the - at the time of the Great Depression much less integrated - 1930s world economy would have allowed for. The issue areas for which financial and economic health is highly relevant have become more complex. States are signified by a degree of interconnectedness and interdependency, both through economic and political channels, that is as yet unprecedented. Scenarios have already been conceptualized that predict a new surge in state capitalism across the globe (Bremmer, 2009) or, on an even more disquieting note, political unrest and related violent conflicts, particularly in sub-Saharan African developing countries (Economist Intelligence Unit, 2009). Strikingly little attention is paid to those least developed countries. Yet, while they might be less integrated into global financial markets than industrialized states, they are likely to be affected through a variety of channels. The reasons for which the impact of an economic downturn in most of the western hemisphere countries on the developing world has so far been widely disregarded in news coverage and public debate will be left open to speculation at this point. However, it pays to evaluate these effects for both normative and matter-of-factly motivations.

Every recession is preceded by a period of boom. And any trough will eventually lead to a peak. One of the questions asked in this paper will therefore be: Would least developed countries have been more or less vulnerable had the United States housing market crashed earlier? In order to give an answer to this question, I will carry out a dynamic examination of economic vulnerability by looking at specific indicators. Before doing so, a brief definition of least developed countries will be given and characteristics of the 2000 to 2007 boom period will be outlined. In the following sections, some issues relating to the subprime mortgage crisis and the concomitant economic turmoil will be considered and the general concept of economic vulnerability will be explained and looked at from different angles. Lastly, both dynamic and static vulnerability for least developed countries will be empirically tested and final conclusions will be drawn.

By following this approach, I wish to direct attention to challenges that the poorest and most needy countries in this world are facing. The G-20 summit, held on April 2, 2009 in London, has taken significant steps towards a reform of the International Monetary Fund and the allocation of additional means to the World Bank. Among the postulates of this leading economies meeting are that “emerging and developing economies should have greater voice in these institutions [the Bretton Woods institutions, author's note]” (G-20, 2009a, section VII.), as well as the replenishing of the Fund's and the World Bank's liquidity by a combined USD 600 billion, of which a minimum of USD 100 billion is intended for direct financial help for the world's poorest economies. In addition to that, the Fund's conditional lending capacity for low income countries has been doubled and an additional USD 6 billion will be available over the course of the next two to three years from the sale of IMF gold reserves and generated surplus income (G-20, 2009b, p. 1).

By acknowledging the severity of the economic distress that has an effect not only on industrial countries in the western hemisphere but also on less developed states, the world's leading powers create a precedent in preventing spillovers from developed countries from causing major disturbances in the developing south. It remains to be seen how effectively the additional disbursements will be distributed and used in practice.

2. Least Developed Countries

Throughout this paper, the focus of analysis will be on the so called least developed countries (LDCs). Because designations of economic underdevelopment are somewhat vague and often used interchangeably, I will first define what is meant by the term LDCs and then move on to an examination of structural features that separate these countries from industrialized nations.

2.1. Definition

It has long been recognized that some countries differ from others not only by their level of per capita income but also by their educational and health care system, structural impediments to economic growth, proneness to natural disasters or political instabilities related to authoritative regimes. An early attempt to capture these features is Morris' “physical quality of life index” (Morris, 1979). It is constructed using infant mortality, literacy and life expectancy conditional on reaching the age one and forming a composite index of the elements.

In order to complement Morris' insights and identify the world's poorest and most vulnerable countries, the United Nations Development Programme (UNDP) has established the widely recognized Human Development Index (HDI) that makes use of three components as well: in addition to life expectancy, the educational attainment of a society (as measured by a weighted average of adult literacy and enrollment rates in different levels of education) and purchasing power adjusted per capita income enter into the index (Ray, 1998). While showing that the issue of underdevelopment cannot simply be broken down on differences in per capita income, in order to identify LDCs some additional criteria are required.

This is achieved by another composite index, developed and triennially reviewed by the Com­mittee for Development Policy (CDP). Besides the Gross National Income (GNI) criterion, the CDP measure includes a Human Assets Index, consisting of health and education indicators, as well as an Economic Vulnerability Index that will be explained in more detail in the following sections (CDP, 2008). Asymmetric thresholds for inclusion and graduation are then established that allow for a classification of least developed countries. For example, in the latest 2006 review, countries eligible for inclusion were to satisfy the condition of a tree-year average GNI below USD 745, while the graduation threshold for the same indicator was at USD 900. One important restriction is also the population size of a country, which must not exceed the upper limit of 75 million. I will refrain from listing the complete set of inclusion and graduation criteria here and refer the reader to the “Handbook on the Least Developed Country Category: Inclusion, Graduation and Special Support Measures”, published by the CDP^[1].

One last remark must be made concerning the sometimes confusing designations of structu­rally weak and poor countries. Due to a variety of institutions that each define their own standards of classification, several notions persist in common use. Among them is the term “developing country”, sometimes referring to the category above least developed countries, but also used in an inclusive sense. Since the expression “developing country” implies a normative assessment of the degree to which a country is able to improve its economic status (which similarly applies to the notion of “less developed countries”), Switzerland now makes use of the term “partner country”. The distinction between North and South is still common, albeit declining in importance; and categorizations into the first, second, third and fourth world are misleading insofar as the fourth world is more commonly defined as consisting of indigenous nations outside of sovereign territorial boundaries (cf. Griggs, 1999).

Throughout this paper, I will use the term “least developed countries” when referring to the classification made by the UN. Most of these countries are situated in sub-Saharan Africa, with approximately one third being in the Asian-Pacific region and one Latin American country (see Annex A for a complete list of countries currently classified as LDCs). Cape Verde graduated from LDC status in 2007 but is still included in the data samples due to inconsistent information provided by the different United Nations organizations. Tuvalu has been excluded from statistical analyses since no data for this country are reported by the World Bank.

As can be expected by the above discussion, there are some commonalities across the scope of least developed countries, some of which will be presented in the next section.

2.2. Structural Features

When economic historian David Landes titled his speech held during the annual meeting of the American Economic Association in 1989 “Why Are We So Rich and They So Poor?” (Landes, 1990, quoted in Jones, 2002), he clearly implied that less developed economies must be set apart from industrialized countries owing to structural features they exhibit.

For one, demographic factors differ to some extent. “Very poor countries”, as Ray notes, “are characterized by both high birth and high death rates” (p. 34). He continues to explain that in the process of economic development, death rates slowly decrease while birth rates stay at an elevated level for a longer period, resulting in high population growth. It is therefore possible that per capita income initially declines even though the country is progressing.

A second feature of least developed countries is their concentration of labor in the agricultural sector. Ray points to the fact that in very poor countries, almost one third of total output comes from agriculture, while at the same time income shares of this sector only account for one to seven percent. Not surprisingly, the attractiveness of moving to urban areas is perceived as being particularly high. This may explain the bulk of migration flows from rural to urban areas, illustrated by the observation that the average rate of urban population growth from 1980 to 1993 was twice as high as the respective rural population growth rate (Ray).

Apart from these properties, differences can also be seen in the composition of exports, a facet that is to be discussed later in greater detail. Given that least developed countries are usually well endowed with natural resources, their being primary commodity exporters comes as no surprise. It is, however, interesting how these exports are composed. Figure 1 shows the share of fuel, food, agricultural raw material and manufactures exports from 2000 to 2006. In sharp contrast to developed economies, where manufactures and consumer durables make up a large portion of export goods, it shows that developing countries rely heavily on crops and food. According to Ray, “primary products are particularly subject to large fluctuations in world prices” (p. 40), indicating that instability in exports could lead to higher vulnerability of the world's poorest countries.

Figure 1: Composition of Merchandise Exports in LDCs

illustration not visible in this excerpt

Before discussing the issue of vulnerability at length in section 5, I will first provide a quick overview of the economic development from 2000 to 2007 and review the causes and effects of the steep economic decline that is associated with the subprime mortgage crisis.

3. The Economic Development from 2000 to 2007

An economic boom period is generally considered to consist of the following ingredients: (1) the presence of expansionary monetary policy leads to low interest rates which in turn stimulate investment; (2) the degree of production capacity utilization increases; (3) inflation tends to rise and (4) unemployment rates are likely to fall. One could add to these factors the concomitant surge in national income and consumption spending.

Many of those elements have been contributing to the upswing of the world economy from 2000 to 2007, as is shown in figure 2. The impressive growth performance of EastAsian economies that temporarily came to a halt during the 1997 Asian Crisis continued to show an upward trend from 1998 on. Inflation as measured by the consumer price index (CPI) rose by almost 30% worldwide and showed an even steeper trend in Latin American and Asian economies as well as in the least developed ones. Finally, real interest rates in the United States fell from almost 7% in 2000 to approximately 1.5% in 2004, before they slowly began to climb again.

illustration not visible in this excerpt

Figure 2: Indicators of Economic Development

Another development that is not directly related to the study of booms and crises, but nonetheless characteristic of recent economic activity, is displayed in the fourth quadrant of the above graph. It shows that trade with other countries constitutes a rather large percentage of the respective country group GDP and that this share is increasing. Especially Asian economies seem to have experienced a surge in trade activity from 2000 on. Yet, since this paper is not devoted to a study of globalization but rather to an examination of crises and economic vulnerability, I will refrain from discussing these developments in greater detail and instead turn to the question whether business cycles have evolved in such a way that specific country groups have become more or less dependent on economic growth in others.

3.1. Decoupling or Convergence?

As the pace of globalization picks up speed, a growing number of scientific articles is concerned with the analysis of global business cycles. Economic convergence, meaning the synchronization of business cycles across different countries or regions, has traditionally been studied either between a similar set of countries or between poor and rich regions in the same state^[2]. The focus has more recently turned to a global approach, incorporating inter-group rather than intra-group analyses and making use of sophisticated econometric techniques in order to account for patterns across a multitude of macroeconomic variables. Before discussing these findings in more detail, let me point to the debate on the issue of business cycle comovement and in particular the competing positions circulating in magazines and newspapers, but also in the scientific community.

In an article headed “Decoupling: Theory vs. reality”, Conrad de Aenlle refers to the term “decoupling” as the “latest big idea to shrink dramatically when tested in the real world” (2008, p. 1). What does this idea stand for? In short, it holds that major emerging economies, but also European countries or other developed economies in general, have become independent from the economic development in the United States to a degree that they are less affected by business cycle fluctuations or even a major recession in North America. The underlying idea is that export structures of countries outside the United States have moved away from a purely inter-group concentration and are instead becoming more intra-group oriented. While de Aenlle continues to exemplify his claim by pointing out the large losses that countries like Japan or Germany have experienced in the aftermath of the subprime mortgage crisis, a shortly afterwards published Economist (2008) article applies itself to a more thorough discussion of the decoupling theory. The argument is made as follows: contrary to the common belief that along with the integration of the world economy, the economic health of one country is closely linked to the well-being of another, growth in income, consumption and investment has been stable or even surged in many emerging markets while the United States already found themselves in an economic downturn. The most convincing evidence for this relationship is that over the course of the last four years, exports from emerging markets to the United States have continually decreased, while those to other countries in the emerging market group have stayed the same (or, for some countries, even risen).

A comprehensive empirical examination of the decoupling vs. convergence question has been attempted by Kose, Otrok and Prasad (2008). They use a set of 106 countries which is divided into industrial countries (INCs), emerging market economies (EMEs) and other developing economies (ODCs). The distinction of the two latter groups is mainly based on their respective degree of trade and financial integration. Data is obtained for a range between 1960 and 2005, allowing for a subclassification of specific time periods, namely the pre-globalization period (1960-1984) and the globalization period (1985-2005). Furthermore, they decompose business cycles into global and group-specific ones. By doing so, they are able to show that volatility of the global factor has decreased from 1.4% in the first period to 0.5% in the second. In order to estimate the effects of global, group- and country-specific factors on variation in output, they use a dynamic factor model. Interestingly, country-specific factors are shown to account for 60% of the variation in EMEs output, while the impact on INCs and ODCs is much lower (39% and 44%, respectively). This leads to the conclusion that “the degree of comovement across the three main macroeconomic aggregates is much greater within countries in this group” (p. 16).

In a second step, the authors compare the pre-globalization and globalization period in order to shed light on the validity of convergence and decoupling hypotheses. Since for all three country groups the average contribution of the global factor to output falls, they reject the former and state that decoupling can be seen to some degree in all three country subsets. However, the most impressive drop is associated with INCs (from 28% to 9%) and EMEs (from 13% to 4%), while other developing countries experienced a decline from 10% to 7%. At the same time, the contribution of the group-specific factor has gained in importance for INCs and EMEs, which is explained as relating to the surge of trade and financial linkages among the respective country groups.

To summarize these results in the words of Kose, Otrok and Prasad, “there has been a substantial convergence of business cycles among industrial economies and among EMEs, but there has also been a concomitant divergence or decoupling of business cycles between these two groups of countries” (p. 23 ff.). Their evaluation also shows that business cycles of developing economies, and, due to their even less pronounced economic ties with industrial economies, least developed countries in particular, have not decoupled from global macroeconomic conditions to a considerable degree, although the comovements are on a generally low level.

The next section applies itself to an exploration of boom's permanent companion, namely the subsequent economic downturn that inevitably forms part of every business cycle. It examines the buildup of the subprime crisis in detail and describes its effects on least developed countries.

4. The Economic Crisis of 2008

The whole wealth so swiftly gathered in the paper values of previous years vanished. The prosperity of millions of American homes had grown upon a gigantic structure of inflated credit, now suddenly proved phantom. Apart from the nation-wide speculation in shares which even the most famous banks had encouraged by easy loans, a vast system of purchase by instalment of houses, furniture, cars, and numberless kinds of household conveniences and indulgences had grown up. All now fell together. The mighty production plants were thrown into confusion and paralysis. But yesterday had been the urgent question of parking the motor-cars in which thousands of artisans and craftsmen were beginning to travel to their daily work. To-day the grievous pangs of falling wages and rising unemployment afflicted the whole community, engaged till this moment in the most active creation of all kinds of desirable articles for the enjoyment of millions. The American banking system was far less solidly based than the British. Twenty thousand local banks suspended payment. The means of exchange of goods and services between man and man was smitten to the ground, and the crash on Wall Street reverberated in modest and rich households alike.

— Sir Winston Churchill, Memoirs of the Second World War, p. 32

Were it not for the perfidious title of the book, one would most certainly be inclined to ascribe the above quote to the economic turmoil that pertains after the breakdown of the subprime mortgage market in the United States and has by now developed to a full-fledged economic downturn. Sir Winston Churchill, in his capacity as British Prime Minister and historian, expresses in his eloquent style what may as well be a barometer of today's society's disposition. Yet he is clearly describing the effects of the stock market crash in 1929, entailing the worst recession of the 20th century and ending a boom period that has oftentimes been called the “Roaring Twenties”. The similarity both in the impact and the roots of the Great Depression following on “Black Friday”^[3] and the recession resulting from the burst of the United States housing bubble in mid 2006 is striking. The subprime mortgage crisis clearly is the dominant phenomenon in news coverage, politics and public opinion. Its effects are widespread: not only is there a general uncertainty within the population regarding the ethical and intellectual quality of the financial sector, but the real economy is by now experiencing a slowdown that makes its effects felt even in remote industrial branches that have at first glance nothing to do with the impressive write-downs on certain banks' balance sheets. Important questions remain to be asked: is greediness of bankers at the heart of the current financial crisis? If not, which causes can be identified and to which degree are they relevant? Where do the effects of the resulting economic turmoil show predominantly?

The following section will outline the anatomy of the subprime mortgage crisis, addressing possible causes as well as bringing the events in a coherent order.

Since the subprime crisis originated in the United States and the deleterious impact of the crisis cocktail consisting of asymmetric information, excessive rent seeking, valuation difficulties and some other potentially explosive ingredients first hit other industrialized countries with highly integrated financial systems, we will first examine the effects on developed economies. It is, how­ever, imprudent and economically shortsighted to leave the far bigger share of developing eco­nomies and in particular least developed countries (LDCs) out of the equation. After an appraisal of the global economic boom period from 2000 to 2007, a review of the facts leading to the build-up of the subprime mortgage crisis and an overview of the effects on the world economy, specific issues concerning LDCs will be examined in more detail.

4.1.Anatomy of the Subprime Mortgage Crisis

By now, the consensus has been established that the subprime crisis originated in the US housing market, spreading through a chain of securitizations on ever more complex layers to banks' balance sheets and finally to the real economy. As one example, Reinhart and Rogoff (2008) express this view by stating that “the United States sub-prime crisis, of course, has its roots in falling U.S. housing prices, which have in turn led to higher default levels particularly among less credit-worthy borrowers. The impact of these defaults has been greatly magnified due to the complex bundling of obligations that was thought to spread risk efficiently. Unfortunately, that innovation also made the resulting instruments extremely nontransparent and illiquid in the face of falling house prices” (p. 3).^[4] However, this consensus refers to where the build-up of the financial crash started and how the crisis unfolded rather than being concise about the why. Indeed, there is much less agreement on the possible causes than there is on the involved markets and institutions both in the academic community and in the political sphere. The following section will present approaches that differ in their emphasis on specific aspects of the crisis while reaching similar conclusions.

According to Clerc (2008), the subprime crisis has its roots in “the combination of historically low interest rates, credit and monetary expansion and booming asset prices” (p. 25). Low interest rates in combination with monetary expansion lead to an expansion in the volume of issued credit, thereby decreasing the incentives for lenders to correctly assess the borrowers' risk. In other words, when more money can be borrowed at cheaper rates, the likelihood of risky investments increases^[5]. Liebowitz (2008) expands on this view in emphasizing the role political institutions have played in the process. He argues that subprime mortgages are ultimately a politically intended instrument to provide for high home ownership rates (especially among ethnic minorities, as will be pointed out later). Evidence suggests that since the foundation of the Federal Housing Administration (FHA), created for the purpose of removing risk from banks by guaranteeing mortgages against default, loan underwriting standards in the United States have continually become more flexible. According to Liebowitz, three major intrusions distorted the workings of the invisible hand in the mortgage market: Fannie Mae, a government-sponsored enterprise (GSE) was founded in 1938 with the initial purpose of purchasing FHA mortgages. The scope of its activities was later expanded to include purchasing and repackaging of a large share of private mortgages in the United States. Second, the Community Reinvestment Act (CRA) required mortgage brokers to “conduct business across the entirety of the geographic areas in which they operated” (p. 6) from 1977 on. Third, the Home Mortgage Disclosure Act (HMDA), passed in 1975, made mortgage lenders provide detailed data on mortgage applications. When in 1992 the Federal Reserve Bank of Boston conducted a statistical analysis using HMDA data on mortgage rejection rates with regard to ethnic origin and “minorities were found to be denied mortgages at higher rates than whites” (Liebowitz, p. 6), the political implications were seemingly clear. In spite of Day's and Liebowitz' (1998) findings that the data used in the study were faulty and did not indicate a minority bias when applied correctly, lending standards were relaxed further so as to encourage subprime lending practices.

This allows us to classify the identified causes into two categories: low interest rates and monetary expansion are economic factors contributing to the bubble in US real estate markets^[6] whereas the credit expansion and the resulting boom in house prices has been deliberately fueled by political actors. It bears repeating that not only economic variables such as the low interest rate favored a faulty risk assessment of the initial mortgages and of the derivatives that followed in the chain of securituzation, but also “a mortgage system run by flexible underwriting standards, which allowed these speculators [i.e. people that bought one or several houses in the anticipation of high rates of return, author's note] to make bets on the housing market with other people's money” (Liebowitz, p. 26).

At the same time, namely from 2000 to 2005, the United States real estate market experienced a strong boom period. House prices rose by an annual average of 8% in nominal terms or 5.5% in real terms, as measured by the Office of Federal Housing Enterprise Oversight (OFHEO) index^[7]. During 2006 and 2007, however, the annual inflation-adjusted growth rate of house prices was virtually zero. Figure 3 graphically shows the development of house prices as measured by the Case-Shiller house price index. The burst of the bubble and the subsequent decline of prices, marked by a vertical red line, can be nicely seen.

Figure 3: Case-Shiller House Price Index

illustration not visible in this excerpt

Still, investments in mortgage-backed securities (MBS) and, on a layer of added complexity, collateralized debt obligations (CDOs), became very attractive “since they were providing higher yields compared to corporate bonds with the same rating and were supposed to be less exposed to the default of any of their components” (Clerc, p. 25). Two other factors contributed to the rising demand for those assets: the declining issuance of government bonds and the global asset shortage that channelled demand from emerging market economies to the United States. This point is illustrated by Caballero, Farhi and Gourinchas (2008), who claim that the “Emerging Market [...] crises at the end of the 1990s, the subsequent rapid growth of China and other East Asian economies, and the associated rise in commodity prices in recent years, reoriented capital flows from emerging markets toward the U.S.” (p. 2 ff.).

The chain of securitization can be reconstructed as follows: credits for houses are offered by banks and specialized companies, taking as collateral the houses themselves. Government sponsored enterprises such as Fannie Mae and Freddie Mac created products that were secured by the mortgage on houses rather than the value of the house itself, so called mortgage backed securities. In a next step these MBS were bundled and sold as collateralized debt obligations or asset backed securities (ABS). As an additional layer of complexity, some companies constructed CDOs of CDOs (also called CDOs squared). When tracing back the underlying assets by way of backward induction of the above chain, it becomes clear that the functioning of the market for these investment products depended on one fundamental variable: the prices for houses. If house prices had risen forever, the system would probably have worked for many years to come. Taking into account that with every additional layer of complexity the model uncertainty rises, Clerc states that “the build-up of domestic and global imbalances was amplified by a double difficulty regarding (1) the fair valuation of assets [...] and (2) risk discrimination” (p. 25). The next chapter will provide a framework for analyzing financial crises with regard to informational asymmetries and valuation aspects.

4.2. ValuationAspects and Asymmetric Information

Some fundamental causes of the subprime mortgage crisis have been identified in the section above. It is important to notice, however, that the degree of damage done to the world financial system would have been far lower had the speculative assets been valued based on market values rather than on expectations derived from mathematical models. Ultimately, the high demand for collateralized debt obligations and asset backed securities can be traced back to the high yields these vehicles provided and to the redirection of financial flows from emerging economies to industrialized countries in search of investment opportunities; albeit this does not constitute the cause of the breakdown.

Banks have had to write down far larger amounts than they actually experienced in credit losses. Estimates for major banks' total losses by December 31, 2008 are at approximately $146 billion, only 18% of which can be attributed to credit losses^[8]. “There is obviously a large amount of uncertainty surrounding the latter figures”, Clerc points out, “as most exposed assets are pretty hard to value in the current juncture” (p. 25). Nevertheless, the impressive write downs on banks' balance sheets, surpassing credit losses by a multiple, clearly imply that asymmetric distribution of information and valuation uncertainties played a major role in determining the magnitude of the financial turmoil. Perraudin and Wu (2008) specifically look at the determinants of asset-backed securities in crisis periods. Their findings are consistent with the hypothesis of valuation uncertainties, as can be seen in Figure 4. It displays home equity loans asset backed securities of different quality and the respective spreads of their residuals. Starting in mid 2007, the dispersion of residuals around a trend value increases vigorously, meaning that from this point in time the prices of ABS cannot solely be explained by fundamental values anymore.

Figure 4: A-Rated Home Equity Loans ABS Fitting Errors

illustration not visible in this excerpt

[...]

---

^[1] An electronic version of this document can be found at

http://www.un.org/esa/policy/devplan/cdppublications/2008cdphandbook.pdf.

^[2] As an example, Barro and Sala-i-Martin (1992) examine convergence in per capita income between the 48 contiguous U.S. States; Quah (1996) is interested in output convergence between different European regions.

^[3] Friday, October 25,1929 was a remarkably tranquil day on Wall Street with the Dow Jones even going up by 1.75 points. Harold James, the Princeton University economic historian, points out that “Black Monday” would better serve its name (James, 2008).

^[4] Gorton (2009) as well as Demyanyk and Van Hemert (2008) describe this development in a similar fashion.

^[5] The underlying assumption for this statement is the presence of asymmetric information. Mishkin illustrates this point in his 1996 paper “Understanding Financial Crises: A Developing Country Perspective”.

^[6] The Central Bank of the United States (Federal Reserve System) is considered to be relatively independent of the government. Thus, decisions taken by the Fed can be seen as being of economic rather than of political nature. However, as it would be foolish to assume that there is no influence of the United States government on its central banking system whatsoever, the established connection is a relative one.

^[7] See http://www.ofheo.gov/hpi_download.aspx for downloadable data on house prices.

^[8] See http://www.stockweb.blogspot.com/2008/02/total-writedown-and-credit-losses-due.html.