Asset Shortages, Global Imbalances and Developing-World Bond Markets

John Hallett

August 26, 2011

Abstract: This paper examines Ricardo Caballero’s novel theory about the role that “asset shortages” play in the contemporary pattern of global capital flows and current-account imbalances. After surveying the existing literature on global imbalances, it relates Caballero’s hypothesis to the developing world’s burgeoning domestic bond markets and investigates whether those markets might serve to alleviate global asset shortages. It concludes with a preliminary econometric investigation of the effect that these debt markets might be having on U.S. interest rates.

I. Introduction

Talk of global current account “imbalances” has become so ubiquitous in academic and popular discussions about the world economy as to verge on the precipice of the cliché. VOX, the economics blog popular among international economists, has a channel dedicated to the topic. The IMF has published a string of papers dedicated to the issue. The G20 has issued statements pledging cooperation on the issue.

Nor do economists and political scientists lack ideas about determining responsibility for the current situation of outsized deficits and surpluses. Nationalist-leaning pundits in surplus nations bemoan the fiscal and monetary profligacy in deficit countries, while their counterparts in debtor nations accuse surplus governments of manipulating their currencies or suppressing domestic consumption.^[1] International bodies call for enhanced monitoring and coordination of fiscal and monetary policies as they relate to imbalances.^[2]

Why so Glum? Reasons for concern about Global Imbalances

The reasons for such a relentless focus on imbalances are numerous. For one, the current pattern of current account deficits seems to contradict traditional economic theory. Classical economics tells us that developing countries, which have high ratios of labor to capital relative to the industrialized world, should import capital from rich countries. With strong future earnings potential, the developing world should be able to make better use of global savings than the capital-rich industrialized world. Instead, as Robert Lucas pointed out in an article that would make his name synonymous with the phenomenon it explored, capital seems to be flowing from the developing world to the rich world.^[3] 

Yawning current account deficits also evoke memories of the financial crises that devastated Asia and Latin America in the 1990s, when the capital flows that had financed those regions’ deficits came to a sudden stop. Some observers have drawn connections between global imbalances and the recent global financial crisis.^[4] Former Treasury secretary Henry Paulson, for example, has said that current-account imbalances depressed global interest rates and thus increased risk appetite in the years running up to the 2008 financial crisis.^[5]

        The United States’ persistent current account deficits have also aroused concern about the potential geopolitical implications of global imbalances. An article published by the Council on Foreign Relations taps into a theme common among Americans fearful of national decline, pointing out that Great Britain’s political and military primacy waned at the same time that it transitioned from creditor to debtor status, and asking whether the dollar might lose its reserve currency status if the United States’ large current account deficits persist.^[6] Indeed, questions about the United States’ deficits’ impact on the dollar looms large in the literature about global imbalances. Kenneth Rogoff estimated in 2007 that a 20 percent decline in the dollar’s value was likely.^[7] Blanchard et. al. estimate that, given a trade deficit of 5 percent of GDP, a net debt-to-GDP ratio of 25 percent, and an interest rate of 4 percent, the dollar would have to depreciate by 40 to 90 percent from its 2005 level.^[8] Since then, of course, the financial crisis and ensuing recession have substantially eroded the fundamentals behind the U.S. economy, increasing the likelihood of a further drop in the dollar’s value.

        Because of the potential impact of global current account imbalances on these three areas of concern – economic efficiency, global financial resilience, and geopolitical stability – it is certainly prudent to investigate the underlying forces behind the imbalances, and to ask whether any recent developments point to a path out of the current situation.

Asset Shortages and Global Imbalances

Ricardo Caballero has proposed an appealingly simple theory that identifies the root cause of the current pattern of global current account imbalances: Global asset shortages. According to Caballero, countries and regions differ in their abilities to produce financial assets that can act as safe stores of value.^[9] Those states with underdeveloped financial markets cannot offer their citizens instruments in which to securely invest in order to save for the future. As a result, to the extent that a country’s demand for safe assets exceeds its ability to produce those assets, money will flow from that country to those states that produce secure and liquid investment vehicles.

Because the United States is the foremost producer of safe, liquid assets – in particular U.S. Treasury bonds – and because those instruments are denominated in the most widely traded currency in the world, it has become the top recipient of global savings. These capital inflows, according to Caballero, have driven the United States’ current account deficits:

The more visible global imbalances [that is, the current account deficits]…were caused by the funding countries’ demand for financial assets in excess of their ability to produce them, but this gap is particularly acute for safe assets since emerging markets have very limited institutional capability to produce these assets. Thus, the excess demand for safe assets from the periphery greatly added to the US economy’s own imbalance […].^[10]

The resulting pattern of capital inflows to the United States allows it to run persistent and large current account deficits. One channel through which these capital inflows facilitate current account deficits is through the inflows’ effects on the cost of borrowing in the United States. Warnock and Warnock (2006) find that during the twelve months to May 2005, official purchases of U.S. treasury notes and government-sponsored agency bonds alone lowered the yield on 10-year Treasuries by 90 basis points, implying that the effects of all capital inflows is substantially larger.^[11] Lower U.S. interest rates strongly influence other interest rates in the U.S. economy, including corporate bonds, mortgage rates, and savings accounts. Lower rates on these instruments, in turn, deter savings while promoting borrowing and consumption.

        Viewing the global financial landscape through the lens of asset shortages, Caballero says, reveals not just the patterns of global current account balances, but also the history of financial bubbles over the past two decades. When the Japanese real estate and stock bubble burst in the 1980s, according to this view, a crucial source of assets was destroyed. With global demand for safe assets still growing, investors simply shifted funds to the United States, which saw widening current account deficits in the 1990s that coincided with the bursting of Japan’s bubbles.^[12] The size of the U.S. financial sector saw a corresponding explosion: U.S. financial assets reached nearly 480 percent of GDP in the third quarter of 2007, up from less than 160 percent in 1980.^[13] According to Caballero, this expansion was driven by global investors’ demand for safe assets.

         Caballero also points to asset shortages as a contributing factor behind the global financial crisis of 2008. The financial crises and currency crashes in Asia and Latin America in the late 1990s and early 2000s had discouraged investors from parking their money in emerging markets. Instead, many of those investors turned to U.S. equities. A few years later, the bursting of the tech bubble and the ensuing NASDAQ crash destroyed another source of assets. As a result, Caballero says, international investors turned their sights to U.S. bonds, the historical object of their affection.^[14]

But the world’s “insatiable demand for safe debt instruments” was far out of proportion to even the United States’ ability to produce such assets.^[15] In 2007 Fitch awarded “AAA” status to less than 1 percent of “single-name” corporate bond issues – that is, to traditional bonds, a category from which Caballero excludes collateralized debt obligations and other derivatives.^[16] In Caballero’s view, this shortage of assets contributed to the marked decline of long-term global interest rates over the past decade, especially in the United States.^[17] This had two important effects with regard to the financial crisis. First, it encouraged borrowing and in turn helped fuel the housing bubble in the United States and Europe. Second, as interest rates fell on traditional debt instruments, pressure grew for the U.S. financial services industry to come up with new products, ushering in the last decade’s securitization boom.

Asset shortages also promote systemic financial risk by encouraging leverage in the financial system, says Caballero.^[18] Banks and nonbank investing houses have a tremendous demand for safe assets with which to back their liabilities. In an environment characterized by shortage, the price of safe assets is high – and yields are correspondingly low. This increases the opportunity cost to banks of backing their liabilities with safe assets. As a result, banks have an incentive to back fewer of their financial obligations, thereby increasing the riskiness of their operations. While Basel III’s proposal to increase banks’ mandatory reserve requirements seeks to address this issue, Caballero is skeptical that such an approach will have a net stabilizing effect on the global financial system, because it generates more demand for safe assets.^[19] 

The financial crisis itself brought a modest contraction of current account imbalances, although US current-account deficits are still large by historical standards (see chart). From Caballero’s perspective, imbalances are likely to persist for as long as the demand for assets in surplus regions dramatically outstrips those regions’ ability to produce safe assets. Surplus nations will continue to export their savings to the United States, fueling its external deficits.

In this sense, the financial crisis did little to alleviate the underlying pressures that have fueled global imbalances. In fact, the crisis exacerbated the phenomenon of global safe-asset shortages to the extent that it destroyed what had become an important category of apparently safe assets: mortgage-backed securities.^[20] As Barry Eichengreen has written, since the crisis, “All that has changed is that foreign central banks are now accumulating U.S. Treasury obligations rather than the securities of government agencies such as Fannie Mae and Freddie Mac.”^[21] Although Eichengreen was referring to official purchases of U.S. securities, the concept is the same when applied to foreign investors more broadly. Thus, surplus nations still park their savings in the United States, propping up the value of the dollar, driving down U.S. interest rates, and financing that nation’s current account deficits.

The only way to alleviate the fundamental forces behind global current account imbalances is for today’s surplus nations to generate enough assets themselves to satisfy at least some of their domestic demand for savings instruments. As Caballero puts it:

The way out of the current juncture is ultimately one of financial development in the regions of the world that have limited capacity to generate store-of-value instruments relative to their demands.^[22]

II. Research Question

This paper will attempt to draw a connection between Caballero’s asset shortages hypothesis and the increasingly numerous surveys of the emerging market economies’ burgeoning domestic debt markets. Although much has been written on both subjects, to my knowledge few authors have elaborated on what global imbalances have to do with emerging-market debt markets and vice versa.

I will also test whether the increasing share of global assets issued by developing countries has an impact on U.S. interest rates. Some authors have conducted empirical investigations as to the effect of financial development on current account balances and national savings rates – I review some of these below – but few have examined the implications that growing emerging-market bond markets might have on the pattern of global capital flows, interest rates, and current account balances.

        The paper is structured as follows. First, I will conduct an overview of the existing body of literature that provides a theoretical link between global imbalances and financial development in the emerging market countries. Second, I will survey other plausible explanations of the pattern of global current account balances to determine whether the current literature can already account for the state of global imbalances, or whether there is room for Caballero’s theory of asset shortages.

Third, I will examine the recent developments in emerging market borrowing, asking whether today’s developing country bond markets are a passing trend or whether something more fundamental is at play. If it is the latter, then these developing bond markets may play a role in mitigating the mismatch between asset demand and asset supply that Caballero says is the main driver behind global imbalances. Finally, I will conduct an econometric analysis of the effect that growing developing-world bond markets might have on U.S. interest rates.

III. Literature Review

The Savings Glut Hypothesis

Caballero’s focus on asset shortages is novel in its emphasis on the supply of safe global financial assets, but it is closely related to a line of thinking commonly known as the “savings glut” argument. In a 2005 speech, then-chairman of the Federal Reserve Alan Greenspan raised the question of why long-term interest rates had persistently fallen despite the monetary tightening cycle that was then underway, calling the conundrum “without precedent in recent U.S. experience.”^[23] The solution to the mystery, he proposed, lay in international patterns: The fact that interest rates were falling across the globe despite U.S. policy, he said, was due to “an excess of intended saving over intended investment.”^[24]

Speaking the same year, then-Federal Reserve Governor Ben Bernanke elaborated on the same point in a speech that popularized the term “savings glut.” He pointed out that developing countries were responsible for most of the increase in global savings over the previous decades.^[25] Federal Reserve figures show that developing countries had moved from a collective deficit of $88 billion in 1996 to a surplus of $205 billion in 2003 – a period over which the U.S. current account deficit increased by $410 billion (Bernanke 2005). The deterioration of the current accounts of France, Italy, Spain, Australia, and the United Kingdom during this time period further suggested that global, rather than just domestic, forces were at work.

The roots of these “excess savings” in the developing world are diverse. Greenspan pointed to the increasing share of the global economy made up by these high-saving nations, as well as to the increase in oil prices over the previous decade.^[26] For Bernanke, the financial crises that struck emerging markets in the 1990s played a crucial role by forcing governments, households and firms to cut back consumption and take precautionary measures to stave off future crises. This included increased personal savings, the accumulation of foreign exchange reserves, and the embrace of export-oriented development strategies. All of these forces increased the savings rate in the developing world, Bernanke said, and shifted current accounts there toward surplus.

Gruber and Kamin (2007) find evidence to support the claim that Asia’s crisis did indeed play a role in shifting the region further toward surplus than the fundamental economic fallout from the crises would have caused alone. Their model employs the standard determinants of current account deficits such as income, growth, budget balances, net foreign asset positions, trade openness, and demographic variables. While these factors fail to explain the full extent of Asian surpluses after 1997, Gruber and Kamin find that financial crisis indicators account for most of the difference.^[27]

Proponents of the savings glut hypothesis have a similar narrative to Caballero as to how these excess savings fueled the U.S. current account deficit. Rapid technological innovation fueled optimism about future productivity gains in the United States, fueling a rise in the price of equities. This increased household wealth and thereby encouraged consumer spending.^[28] The same capital inflows that pushed up the value of the stock market also bid up the value of the dollar, boosting imports and hurting exporters’ competitiveness. Both of these factors were compounded by the large stock of savings in the developing world, and both pushed the U.S. current account deficit downward.

Those increasing asset prices, of course, turned into the NASDAQ bubble. Once equity prices entered its correction phase, new investment declined and demand for financing was substantially reduced, “yet desired global saving remained strong.”^[29] Textbook economics tells us that with investment demand falling and savings still high, interest rates will fall. Viewed through the lens of the savings glut hypothesis, it is a sign of just how far savings outstripped investment that rates continued to fall even once the economy recovered and the Federal Reserve initiated a tightening cycle.

These low interest rates altered the “transmission mechanism” by which the global savings glut fueled the U.S. current account deficit. Instead of fueling a stock-market bubble, Bernanke said, global savings held down long-term interest rates. These low interest rates fueled the housing bubble of the 2000s, with a similar wealth effect as the stock market boom of several years earlier. Homeowners used the rising values of their homes to increase their expenditures: Four-fifths of the rise in mortgage debt in the three years to 2005 was due to discretionary extraction of home equity.^[30] Half of that equity extraction fueled household expenditures, split equally between direct expenditures and financing existing household debt.^[31] The overall result was a further deteriorating U.S. current account, financed by foreign savings.

The fact that savings from the developing world are financing consumption in the rich world, according to proponents of the savings glut hypothesis, is due to financial underdevelopment in the former.^[32] This leaves savers in the developing world with few options if they want to invest their funds locally. Due to the widespread opening of capital accounts in the 1990s and 2000s, though, savings are freer than ever to find investment opportunities overseas. As Mendoza et al (2007) put it, “Financial integration was a global phenomenon, but financial development was not.”^[33] This situation “encourages countries with excess savings to seek financial intermediation in well-developed financial systems such as the United States.”^[34]

The evidence is mixed as to whether a “global savings glut” actually exists. On the one hand, world savings did increase modestly from the 1990s to the 2000s, rising from an average of 22.6 percent of gross world output in the period 1987-1994 to 24.4 percent in 2003-2007.^[35] In China, as in other East Asian nations, gross savings was led by the high savings rate of the corporate sector, which accounted for nearly half of Chinese savings in the 2000s.^[36] This outpaced even the rapid increase in investment there, leading to a growing net savings rate.

On the other hand, Obstfeld and Rogoff find that most of this increase in global savings started only in 2004 and occurred largely after Bernanke put forward his hypothesis in 2005.^[37] Prior to that, they find, the lack of investment demand after the NASDAQ crash, rather than rising global savings, accounted for most of the decline in long-term global interest rates. “It is only around 2004 that the idea of a global savings glut (as opposed to a global dearth of investment) becomes most plausible,” they write.^[38]

Other Explanations

A host of other explanations look to domestic policies, rather than global phenomena, as the main drivers of global imbalances. One focuses on monetary policy. Alan Greenspan believes that the decline in long-term interest rates that has characterized global imbalances could be the result of lower inflation expectations, a reduced risk premium from lower inflation volatility, and a low term premium due to the alleged moderation of the business cycle.^[39] Warnock and Warnock (2006) believe that the most important factor behind the decline in long-term interest rates since the 1980s is the improved credibility of the Federal Reserve, especially since the Greenspan era.^[40]

Others who attribute U.S. deficits to monetary factors paint a less flattering picture of U.S. monetary policy. XXXXXX’s empirical work suggests that monetary shocks, rather than savings shocks, are the key drivers of current account balances in developed and developing countries alike (ECB Savings Glut p. 17). They point to negative real interest rates in the United States from 2003 to 2005 as the primary culprits behind the burgeoning U.S. current account deficits during that period (ECB Savings Glut p. 8). Similarly, Chinn writes that U.S. monetary policy directly fueled that country’s housing boom, the wealth effects of which served to dramatically increase consumption and reduce saving.^[41]

Further characteristics of the U.S. economy might also help to drive global imbalances as well. Oil, which makes up a substantial portion of the U.S. trade deficit, has risen steeply in price over the past decade. Chinn blames federal policy for doing little to reduce the United States’ dependence on imported energy and points out that the U.S. tax rate on gasoline is lower in real terms than it was before the first oil crisis (Chinn, “Getting Serious About the Twin Deficits” p. 9). On a more basic level, responsiveness of U.S. imports to U.S. income exceeds the responsiveness of U.S. exports to foreign income, leading U.S. imports to grow faster than exports if the United States grows at the same rate as its trading partners – and the United States has grown faster than its most important export market, Europe.^[42]

        Perhaps the most prominent school of thought that views global imbalances as a “made-in-America” phenomenon is the “twin deficits” hypothesis, so named because of the simultaneous expansion of both the federal budget deficit and the current account deficit in the United States during the Reagan administration. Federal budget deficits, argue proponents of this view, have negated high corporate savings during the past decade to create an “American-made savings drought” that has sucked in capital from abroad to finance U.S. current account deficits.^[43] In particular, Chinn argues that tax cuts and increased government spending during the Bush administration boosted economic activity and drove private consumption.^[44] Furthermore, he writes, the budget deficits increased interest rates to a level higher than they would normally be, which attracted international investment, bid up the price of the dollar, and hurt exporting industries.

Experience has shown that net government spending does in fact affect national savings rates, in contrast to what the theory of Ricardian equivalence tell us;^[45] however, it is a matter of debate as to how much reining in the U.S. budget deficit would affect the U.S. savings rate. One study suggests that a one-dollar reduction in the federal budget deficit would cause the current account deficit to decline less than 20 cents.^[46] Chinn, writing in 2005, estimates that number to be closer to 33 cents for the United States.^[47] Eichengreen et al (2011) estimate that the U.S. current account deficit will contract by just 15 cents for every dollar eliminated from net federal spending today.^[48] They note that most estimates find the elasticity of current account balances to budget balances to be in the range of 0.2 to 0.3 globally.^[49] Moreover, the effect that budget balances have on current account balances varies across time. In the case of the United States, Eichengreen et al find little evidence that the “twin deficits” hypothesis applies beyond the mid-1980s and

2001-2004.^[50] Gruber and Kamin find that budget deficits did not appear to drive the U.S. current account deficit during the period from 1997 to 2003.^[51]

        Beyond this time period, the twin deficits view has trouble linking budget deficits to the current account. The U.S. current account deteriorated by $300 billion between 1996 and 2000, at the same time that the federal budget came into surplus.^[52] Nor can other hypotheses focused on domestic factors account for global trends like declining interest rates. “If the dominant impulse explaining global events was declining U.S. savings, one would expect abnormally high real interest rates, as with the twin deficits in the 1980s, not abnormally low real interest rates,” Lawrence Summers said in 2006.^[53]

While domestic factors including loose U.S fiscal and monetary policies, reliance on imported oil, strong economic growth, and high propensity to import have certainly all contributed to the pattern of global imbalances over the past decade, none can fully account for the trend. Nor can existing models of savings rates and financial development across countries: Studies by Ito and Chinn, Eichengreen et al, and Gruber and Kamin all consistently under-predict U.S. current account deficits. There is thus room for further investigation as to the causes of the counterintuitive flow of capital from the developing world to the United States and the global imbalances related to that phenomenon.

Can Financial Development Alleviate Global Imbalances?

        A key question raised by Caballero’s asset-shortages hypothesis – and by the savings glut hypothesis more broadly – is whether financial development in developing economies will reverse the current trend of capital flows that go from the developing to the industrialized world. There are several ways that financial development could achieve this in theory.

        First, countries with more advanced financial systems will by definition produce more domestic savings and investment instruments. Because Caballero, Bernanke, and other “savings glut” proponents posit that international savings find their way to the United States for lack of suitable domestic investment options, many developing world savers would no longer be compelled to invest in U.S. assets were their home countries to undergo substantial enough financial development. As Burger et al phrase it, “With more investment vehicles at home, wealth in emerging economies might be less likely to flow to advanced economies.”^[54] Viewed through Caballero’s lens, this would reduce a country’s net demand for safe assets while simultaneously absorbing some amount of world asset demand. This shift in financial flows would reduce current account surpluses in countries that undergo financial development.

        Second, financial development could reduce gross savings. As Bernanke proposes, and as Gruber and Kamin (2007) find evidence to support, East Asia’s financial crisis ushered in an era of current account surpluses in part by spurring firms, households, and governments to accumulate precautionary savings. Households and firms in countries with robust financial systems can smooth out their income by borrowing when the need for cash arises, thereby reducing the imperative for precautionary savings.^[55] Additionally, a well-developed financial system would be able to offer savers instruments with higher rates of return than they have currently. Obstfeld and Rogoff wager that this would result in “huge positive income effects [that] would in all likelihood swamp substitution effects, resulting in lower, not higher, household saving.”^[56]

        Third, even if financial development fails to reduce gross savings, a domestic system of financial intermediation could reduce net savings by facilitating domestic investment opportunities. If domestic investment rises faster than domestic savings, the current account will deteriorate.

        The evidence regarding financial development’s effects on savings rates and current account balances is mixed. Chinn and Ito (2007) fail to find compelling evidence that emerging market economies, especially those in East Asia, would save less if their financial systems were to experience further financial and legal development. They even find some evidence that financial development would increase savings there.^[57] Even when Ito and Chinn (2009) add additional measures of financial development – including financial industry development and equity, bond and insurance market activity – they find stock market capitalization and credit market size to be positively correlated with the current account balance in the developing world.^[58] This would indicate that financial development could actually increase developing world surpluses. All of this suggests that further financial development alone will not reduce current account surpluses for most developing countries.

        On the other hand, Chinn and Ito estimate that financial development led to a decrease in net savings in China over the period 1996-2004.^[59] Although they find that financial development boosted gross savings by 1.7 percentage points, it also increased investment by 2.4 percentage points. The net effect of financial development, then, was a deterioration of the current account.

In addition, Ito and Chinn (2009) determine that financial openness leads to smaller current account balances in developing countries.^[60] Eichengreen et al (2011) find that if China were to increase its financial openness to the level that Thailand had achieved in 2008, it would experience strong capital inflows that would result in a sizeable current account adjustment, even without any change in exchange-rate policy.^[61] The interaction term between capital account openness and financial development was also significantly and negatively correlated with the current account balance. Evidence also indicates that developing countries with robust legal systems experience current account deterioration as a result of financial development.^[62]

These findings suggest that global investors are willing to park some of their savings in developing countries, and that developing countries could absorb some of the world’s demand for assets if they open themselves up to global capital inflows. Financial and legal development would enhance these open countries’ ability to absorb global savings. As Chinn and Ito write, “The results do seem to bolster Bernanke’s argument that more financial development will solve the issue of the savings glut in emerging market countries with high levels of both legal development and financial openness.”^[63]

IV. Stylized Facts on Developing-World Bond Market Growth

Developing World Bond Markets: A Solution to the Asset-Shortages Conundrum?

This section will examine one of the recent trends in financial market development in emerging markets: The rapid growth of domestic bond markets in the developing world. It will first explore the possible ways in which this development could help reduce global imbalances. It will then provide an overview of the scope of these markets’ development and the extent of foreign participation in them. The next section will build on this overview to conduct an econometric analysis of the effect they may be having on U.S. interest rates.

Domestic bond markets in the developing world could address global imbalances on a number of fronts. First, they provide a new source of assets for both domestic and international savers. Caballero hypothesizes that the developing world’s inability to produce sufficient suitable financial assets itself drives its savings into U.S. assets, providing the primary underlying force behind global imbalances.^[64] By reducing this gap between domestic asset demand and domestic asset supply, these new bond markets could reduce the net capital flows from the developing world to the industrialized world. Domestic savers will be attracted to local-currency assets both because of home bias, and firms and pension funds may prefer to hold assets in the same currency as their liabilities.

Second, they could attract international investors looking to diversify their portfolios, increase their exposure to rapidly growing economies, and take advantage of rising currencies. This would increase net capital flows to these developing economies.

        Third, local currency asset markets might reduce gross savings by reducing the need for some precautionary savings. If savers for retirement can invest in high-quality local-currency assets to match their local-currency liabilities, they might not feel compelled to set aside quite as much money to guard against currency swings.

        Fourth, more robust domestic bond markets could reduce net savings by facilitating domestic investment opportunities. Small and medium-size companies that may not have been willing or able to issue international or dollar-denominated bonds in the past might be more likely to issue local-currency bonds on the domestic markets in order to finance capital investments.

        Finally, if local currency instruments in the developing world act as substitutes for the dollar-denominated instruments of the past, the demand for dollars will fall. As the dollar depreciates against other currencies, U.S. imports will fall and exports will become more attractive on the global market.

        Determining the precise extent to which each of these channels impacts global imbalances is beyond the scope of this paper. This section will, however, attempt to provide a starting point for deeper investigation by providing a survey of recent local-currency bond market developments with an eye to its sustainability and its ability to attract foreign investors and conduct a preliminary investigation of the extent to which market growth may already be affecting current-account imbalances.

Scope of Recent Development

Bond issuance in developing countries has expanded rapidly in the past decade, as shown in the chart below, which plots the growth of outstanding domestic and international debt from developing countries. Of particular note is the growth in domestic debt, which, excluding China, reached a level of approximately $4.8 trillion outstanding in 2010. Including China, that figure rises to $7.8 trillion.

The trend shows no signs of abating. In January, the Colombian power firm Emgesa found that demand for its $400 million issuance of Colombian peso bonds outstripped supply by a factor of three.^[65] The next month, the Russian government more than doubled a bond issue originally worth $1.4 billion due to strong investor demand.^[66]

The Russian issue, which matures in 2018, also serves as a testament to the ability of developing-world issuers to borrow for longer terms than was previously common. In fact, a Barclays analysis determines that many 10-year bonds in emerging markets have become so popular that they no longer represent a good value for investors based on real returns.^[67]

Another striking trend is the proliferation of bonds denominated in local currencies. In 2009, 48 percent of government debt in emerging markets was denominated in foreign currency, down from 55 percent just seven years earlier.^[68] Latin America has shown the strongest improvement in the area of local-currency debt, with over 70 percent of its debt denominated in local currency in 2008, up from about half in 2001, while the average maturity has lengthened.^[69]

Foreign Investor Participation

Caballero hypothesizes that financial development in the developing world will reduce the gap between those countries’ demand for assets and their ability to supply them, thereby reducing their need to send their savings to rich countries for financial intermediation. This would shrink emerging-market current account surpluses by reducing gross capital outflows.

Financial development might also reduce global imbalances through another channel: The attraction of foreign funds. If developing countries can produce financial assets that foreigners are willing to invest in, they will experience larger gross capital inflows, leading to lower net capital outflows and correspondingly weaker current accounts. Thus, as Burger et al point out, “One way to evaluate whether local bond markets might play an important role in alleviating the global asset shortage is to analyze the participation of foreign investors.”^[70]

        Foreign investor participation in and of itself can catalyze the development of local bond markets, further enhancing those markets’ appeal. One way foreign investment helps is by developing liquidity. Domestic institutional investors, which make up a large segment of the investor base in many emerging markets, typically hold bonds to maturity.^[71] While this is desirable in the sense that these institutional investors will not flee the market in times of stress, it does not facilitate what Philip Turner calls “micro-liquidity” – the daily turnover needed to reduce market volatility and narrow bid-ask spreads.^[72] Foreign holders, in contrast, usually have higher turnover, which helps to create a liquid market.^[73]

        A prime example of how foreign investor participation has exerted a “snowball effect” on domestic bond markets is the recent growth of emerging-market domestic bond indices. In response to growing investor demand for domestic bonds from the developing world, wealth management funds have begun to incorporate local-currency emerging-market bonds in their benchmark bond indices, thereby improving liquidity in those markets.^[74] This encourages new funds to follow suit. For example, an ETF created by Van Eck explicitly seeks to mirror JPMorgan’s GBI-EMG Core Index, which is comprised of investable emerging-market local-currency bonds.^[75]

Van Eck is not the only asset management firm riding the wave of domestic bond market development in emerging markets. Dreyfus’ sovereign local-currency bond fund grew from $25 million from its foundation in mid-2009 to $1.7 billion as of end-January 2011.^[76] SPDR Barclays Capital launched a new local-currency bond ETF in May, attracting over $50 million by the end of June.^[77] State Street Global Advisors also launched an emerging-markets local-currency bond ETF in Europe in 2011.

These new funds have capitalized on a broader surge in interest in domestic developing country debt. Non-resident holdings of bonds from nine emerging-market governments increased seven fold from 2002 to 2006.^[78] More recently, emerging-market local-currency bond funds saw their biggest inflows ever in the fourth quarter of 2010, with $11.4 billion, for a total of $53 billion of net new capital entering such funds last year.^[79] Local-currency bond funds had $77 billion under management as of May of this year, more than three times as much as at year-end 2009.^[80] The second quarter of 2011 saw local-currency emerging-market bond funds take in twice as much money as did funds specializing in “hard-currency” bonds.^[81] In one week in early June, $410 million flowed into local-currency bond funds, compared with $55 million into hard currency bonds and $87 million going into blended funds.^[82]

        The fact that major institutions like JPMorgan, Barclays, and State Street have a hand in developing-country domestic bonds bodes well for continued foreign involvement in these markets and indicates that trend-driving global investors view them as a valid and attractive asset class. The proliferation of EFTs specializing in local bond markets also promises to attract retail investors to emerging-market securities.

        Both institutional and retail investors have been attracted by high nominal and real yields. During the past decade, higher portfolio concentrations of emerging-market bonds have resulted in higher returns for any given risk level, with volatility comparable to rich-world bonds and low correlation with the U.S. market.^[83] Strong expected currency appreciation in the developing world has added to the attraction. Central banks from Brazil to South Korea have intervened to stem the rise in their currencies’ value in the face of strong recent capital inflows, but market analysts expect these measures to simply delay, and not prevent, the rise of emerging-market currencies.^[84] This is one reason why ten-year emerging-market bonds have retained their popularity despite low real returns – investors believe that upward pressure on currencies is simply being pushed to the back end of the yield curve.^[85] A battery of research suggests that investors are correct in this assertion.^[86]

Although the developing world has seen episodes of strong capital inflows before, there are several reasons to believe that emerging-market domestic bond markets have achieved a “critical mass” and will serve as a reliable source of assets in the future. First, emerging markets are rapidly increasing their share of global world output, and it is only logical for this fact to be reflected in international capital markets. Second, some investors have signaled that they are open to currency diversification in the face of doubts about the major currencies’ strength – or, in the case of the euro, even its continued existence.^[87],^[88] As Charles Robertson, global chief economist of Renaissance Capital, told the Wall Street Journal, “‘The shift began in 2005, but has really accelerated in the last year and a half with investors seeking local currency debt because they don’t trust the dollar anymore, they don’t trust the euro anymore.’”^[89] 

Third, the prospect of slow growth in the rich world for the foreseeable future contributes further to the appeal of local bond markets in the developing world. A recent Barclays report on emerging markets bonds sees “strong signals that a structural portfolio reallocation into EM assets stands behind current cyclical trends,”^[90] noting that “Inflows into EM funds are a natural implication of the two-speed recovery, with EM leading the recovery and the developed world lagging due to fiscal challenges and deleveraging.”^[91]

        Finally, institutional reforms in a number of emerging markets have fostered domestic bond market growth and created a more welcoming environment for foreign investment. Pension fund privatization, for example, has helped to created an investor base with a strong demand for local bonds.^[92] Better inflation performance has helped to establish benchmark rates and instruments, while stronger rule of law has bolstered creditor rights and deepened credit markets.^[93] A falling reliance on adjustable-rate debt securities attests to creditors’ heightened confidence about inflation-fighting policies in emerging markets: Fixed rate instruments represented 71 percent of all outstanding debt instruments in 2007, up from 65 percent in 2002.^[94] Reforms will also have contributed to lower bond yields. One study finds that a 10 percent drop in a country’s perceived economic risk leads to a reduction of 52 basis points in corporate bond spreads for that country’s firms.^[95] 

V. A Preliminary Econometric Investigation

Methodology

         In the analysis below, I regress U.S. interest rates on a novel and simple measure of developing-country financial development: The share of total outstanding debt instruments issued by developing countries. I use quarterly data from the first quarter of 1994 through the second quarter of 2005, due to constraints on data availability beyond that period. The model I use is adapted from the one created by Warnock and Warnock (2006). In that analysis, the authors investigate the effects that foreign official purchases of U.S. government and agency bonds have on the ten-year U.S. Treasury interest rate and other key rates, using a wide range of control variables, most of which I retain in my analysis. Warnock and Warnock find that foreign official flows have a substantial impact on U.S. interest rates, lowering ten-year Treasury yields by 90 basis points during the sample period.^[96] As discussed above, if financial development in the developing world allows households there to keep more of their savings in domestic instruments, less money would flow to the United States and, ceteris paribus, we should see yields rise on U.S. debt instruments. Similarly, if growing debt markets in the developing world compete with U.S. debt issuers for capital from third-party nations, we would expect to see U.S. interest rates rise.

Dependent Variable: Ten-Year U.S. Treasury Interest Rate

        Other studies, including those by Chinn and Ito (2007), Eichengreen et al (2011), Gruber and Kamin (2007), and Ito and Chinn (2009), test whether financial development impacts savings rates and current account balances in developing countries. I have found none, however, that investigate the direct effects that financial development in developing countries has on U.S. interest rates.

Using U.S. interest rates as the dependent variable – as opposed to directly looking at the U.S. current account balance, which is after all the final object of interest – has value for two reasons. First, according to both Caballero and Bernanke, the interest rate is the primary mechanism by which capital flows to the United States impact the U.S. current account. Lower interest rates encourage borrowing, and thereby consumption and imports, and discourage saving. The ten-year borrowing rate for the U.S. government is particularly important because it serves as the basis for a gamut of other key interest rates in the U.S. economy that influence household, commercial, and possibly public consumption patterns.

Second, there are a host of factors that seem to influence current account balances, many of which I have surveyed above. This makes it difficult to judge the effect that any one variable has on the current account, which helps to explain the vigorous debate surrounding the issue of global imbalances. By looking at the mechanism by which developing-world financial development allegedly impacts the U.S. current account, we might be able to get a clearer picture of the forces at work.

The data on interest rates, as well as the control variables, come from Warnock and Warnock (2006), who compile monthly data on several key U.S. interest rates. The data on global debt securities comes from the Bank for International Settlements’ Securities Statistics series, specifically Table 14B (“International debt securities by residence of issuer – bonds and notes”) and Table 16A (“Domestic debt securities by sector and residence of issuer – all issuers and governments”). Because I have employed quarterly data in my regressions, I average the monthly interest rates to come up with quarterly values.

Independent Variable: Share of Global Assets Outstanding

Caballero’s hypothesis about asset imbalances can be rephrased to say that the industrialized world receives a disproportionate share of global capital flows because it produces a disproportionate share of global assets. That is, if the developing world were to produce financial assets in a quantity more in line with its asset demand, it would send less of its savings to the United States. The developing world’s assets as a proportion of total global assets can serve to proxy the developing world’s asset supply and demand mismatch.

Additionally, if the growth in asset issuance over the past decade and a half has been largely demand-driven, as Caballero suggests, then the proportion of outstanding global debt made up of debt from developing countries serves as a measure of how much global asset demand the developing world is absorbing. As its share of total assets rises, the developing world has less unmet demand for assets, and it will send correspondingly less money abroad for financial intermediation. These reduced capital flows should result in higher U.S. interest rates.

Analyzing this aspect of financial development could also measure the competition for capital faced by the United States. As debt from the United States makes up a smaller share of total outstanding debt, interest rates should be bid up as global investors have more options as to where to park their money.

        As the chart below shows, debt from the developing world has grown as a share of total outstanding debt. It rose sharply in the mid-1990s before falling around the time of the Asian financial crisis and staying mostly flat for the next several years as cries in Russia, Turkey, and Latin America scared away global investors and forced deleveraging in developing countries. The middle of last decade again saw a steep rise in developing world debt issuance. Indeed, it is striking that the developing world’s share of global debt increased even amidst the CDO frenzy and widening federal budget deficits in the United States. After a sharp but brief fall in late 2008 and early 2009, growth bounced back and now seems to have resumed its pre-crisis trend.

Data on debt issuance comes from the Bank for International Settlements. Reporting countries labeled by the BIS as “developing” are Albania, Argentina, Armenia, Bangladesh, Bolivia, Brazil, Bulgaria, Chile, China, Chinese Taipei, Colombia, Costa Rica, Croatia, Czech Republic, Egypt, Georgia, Guatemala, Honduras, Hungary, India, Indonesia, Kenya, Lebanon, Lithuania, Malaysia, Mauritius, Mexico, Moldova, Morocco, Nepal, Pakistan, Peru, Philippines, Poland, Romania, Russia, Seychelles, South Africa, Sri Lanka, Tanzania, Thailand, Tonga, Turkey, and Venezuela.

I exclude China from my tally of domestic developing-world bonds outstanding for three reasons. First, strict limits on China’s capital account and current account mean that households and firms already have little choice but to invest in domestic Chinese securities. Second, China’s substantial official purchases of U.S. securities are largely driven by exchange-rate policy and are unlikely to slow down in response to increased investment opportunities at home. Third, sterilization bonds have been a key driver of bond market growth in China. Morgan Stanley estimates that a 2007 issue by the Ministry of Finance represented nearly 15 percent of all outstanding bonds in China at the time, and 80 percent of all government and policy-bank debt in 2006.^[97] These three factors make China an outlier in terms of the forces behind both its bond market growth and its capital flows to the United States.

Control Variables

        The control variables come from Warnock and Warnock (2006), who use standard macroeconomic determinants of U.S. interest rates including inflation expectations, growth expectations, the federal funds rate, a risk premium (which they proxy with the volatility of the ten-year interest rate), and the structural budget deficit as a share of GDP (lagged one year).

In addition to the controls employed by Warnock and Warnock, I add a measure of global liquidity, growth of M3 in OECD countries. This is roughly in keeping with Bracke and Fidora (2008), who measure liquidity in individual countries as the sum of “money” and “quasi-money” from the IMF, and who find liquidity to be a key determinant of interest rates. It is also inspired by Baks and Kramer (1999), who measure global liquidity using monetary growth in G7 countries, and by IMF (2010), which uses G4 monetary growth and finds global liquidity to be a significant driver of asset prices in emerging economies.^[98] In using monetary growth from the entire OECD, I employ a broader measure than Baks and Kramer in order to capture the effect of monetary growth in countries beyond the rich core of the G7. Quarterly comprehensive data on monetary aggregates is not available for many developing countries.

Results

Table 1 shows the results of regressing interest rates on the control variables utilized by Warnock and Warnock: Ten-year inflation expectations (ph_inf10), ten-year inflation expectations minus one-year inflation expectations (dexppi), one-year-ahead growth expectations (bc_y), ten-year Treasury interest-rate volatility (vol_i10), the federal funds rate (iff), and the lagged structural federal budget deficit as a proportion of GDP (budstr_gdp). The control variables alone have a great deal of predictive power, with an R-squared of 0.878.

Table 1: Specification I

The specification shown in Table 2 adds the measure of outstanding developing-world debt securities as a proportion of global outstanding debt securities. While devshare fails to achieve a 95 percent significance level, it is significant at the 90 percent level. The sign on the coefficient is positive, which is the expected direction. This indicates that as developing-world debt makes up a larger share of global total outstanding debt, U.S. interest rates will rise. This specification increases the R-squared only marginally, from 0.878 to 0.886, despite the addition of another explanatory variable.

Table 2: Specification II

Specification III, shown in Table 3, adds OECD M3 growth to the model. While the variable fails to achieve statistical significance, the sign on the coefficient is in the expected negative direction, and the R-squared rises to 0.892 – again, a small rise for adding another variable.

Table 3: Specification III

        Specification IV removes the variables for one-year-ahead growth expectations and the risk premium, both of which registered as highly insignificant in the previous specification. The results are shown below in Table 4. In this specification, the significance of devshare comes close to achieving significance at the 95 percent level, although it fails to cross the threshold. Also of note is the fact that M3 growth achieves significance at the 95 percent significance level, while the R-squared reaches its highest value so far, at 0.890.

The fact that this specification balances the greatest explanatory power with the fewest variables makes it the best fit of the models tried so far. A Durbin-Watson test returned a p-value of 2.89, and a Box-Ljung Q test provided a p-value score of 0.00 and a Q-statistic of 95.59, indicating minimal autocorrelation in the residuals and thus reliable standard errors on our variables’ coefficients.

Table 4: Specification IV

These results suggest that as developing world assets make up a larger share of total global assets, interest rates in the United States will rise. The coefficient on devshare in specification IV indicates that for a one-percentage-point increase in developing-world debt instruments as a share of total global outstanding debt instruments (in this case, 0.01 units), the interest-rate on ten-year U.S. Treasury bonds will rise by nearly 38 basis points – a huge impact. Indeed, developing-world debt made up just 3.3 percent of total outstanding debt at the beginning of 1994. That number rose to 6.3 percent at year-end 2010. This model indicates that this change has coincided with a rise of nearly 113 basis points in ten-year U.S. Treasury interest rates from the level they would otherwise be.

Interestingly, including the United States’ outstanding debt as a share of the global total (usa_share) in the place of devshare fails to achieves significance at even the 10 percent confidence level. The fact that usa_share returns no significant results while devshare does seems counterintuitive at first. It would seem that the same mechanism whereby a higher devshare is correlated with higher U.S. interest rates should also increase rates in the case of a lower usa_share – they seem to be two sides of the same coin. It could, however, indicate that U.S. interest rates are not particularly sensitive to competition from increased debt issuance in, say, Europe or Japan. Rather, savers in the developing world may strongly prefer to hold U.S. assets if they cannot hold domestic assets, and it is only when they are presented with domestic investment vehicles that they move out of U.S. assets. The fact that the sharp rise in outstanding developing-world debt is driven largely by domestic securities – which the BIS describes as “those issued on local market, in domestic currency, targeted to resident investors” – is consistent with this line of reasoning.  

In interpreting these findings, several caveats are called for. The first, of course, is that correlation does not imply causality. It could be that firms and governments in the developing world have taken advantage of low global interest rates – proxied by low U.S. interest rates – to issue more debt. Similarly, the growth in developing-world debt issuance and the decline of U.S. interest rates could be correlated with some third factor not included here. Candidates include a better proxy of global risk appetite, the perceived moderation of global business cycles, and political stabilization in the developing world. Perhaps different measures of global liquidity could have accounted for both phenomena.

Second, devshare never actually achieves statistical significance at the 95 percent confidence level, although it does come close. Third, although Specification IV does return a higher R-squared than Specification I with the same number of variables, it has only marginally more explanatory power. Thus, although the coefficient on devshare is large, it seems to lack much predictive power when combined with traditional macroeconomic indicators.

Fourth, the results are not especially robust to alternative specifications. P-scores fluctuated from one specification to another, and usa_share failed to come close to statistical significance at a reasonable confidence level. As discussed above, this may have important implications for the way that developing-world financial growth affects U.S. interest rates. Alternatively, it could be a sign of the weakness of this model.

Nonetheless, the results serve as a preliminary indication that Caballero’s theory of asset shortages may hold some truth. The predicted effect of the developing world’s increased share of global outstanding debt instruments is sizeable and imply that the United States may lose a key source of cross-border financing if rapid financial development in poor nations releases savers there from the “captivity” of the U.S. financial market. Further econometric investigations that employ data from a wider timeframe and use different measures of liquidity and risk appetite would provide these findings with a more rigorous robustness check and could shed further light on the direction of causality at work here.

VI. Conclusion

This paper has focused on Ricardo Caballero’s theory of asset shortages. It related his writings to the existing literature, including the “savings glut” hypothesis to which it is closely related, as well as works that focus on domestic causes of U.S. current account deficits. It found that U.S-centered explanations of global imbalances fail to fully account for the global pattern of capital flows and falling interest rates that have characterized the past decade and a half. It also concluded that while results of previous studies have so far proved inconclusive as to whether financial development leads to lower savings and deteriorating current accounts in the developing world, there exists strong theoretical and empirical evidence that alleviating the asset shortages identified by Caballero could go some way to reducing global current account imbalances. As such, it concluded that Caballero’s asset-shortage hypothesis offers a promising lead to researchers engaged with the issue of global imbalances.

The paper then turned to the question of whether the striking recent growth in domestic bond markets in developing countries could represent a viable new source of investment vehicles. It found rapidly growing interest on the part of foreign investors in these markets. It also found evidence to support the notion that this interest is built on the back of economic fundamentals and structural shifts in the global financial and economic landscape.

Finally, this paper presented a preliminary econometric test of Caballero’s asset shortages hypothesis by examining the correlation between U.S. interest rates and the share of the global debt market comprised of instruments issued by developing countries. It found that these two factors were strongly and positively correlated, albeit at a confidence level below 95 percent. Further work is needed to conclusively determine the relationship between these two variables. If the results found here hold up to further econometric analysis, they would suggest that financial development in the developing world will result in those countries sending less of their savings to the United States for financial intermediation, consistent with Caballero’s hypothesis.

Works Cited

Bernanke, Ben. "Remarks by Governor Ben. S. Bernanke at the Sandridge Lecture, Virginia Association of Economists: April 14." Richmond, VA, 2005.

Binici, Mahir, Michael Hutchison, and Martin Schindler. "Controlling Capital? Legal Restrictions and the Asset Composition of International Financial Flows." Journal of International Money and Finance 29, no. 4 (2010): 666-84.

Blanchard, Olivier, Francesco Giavazzi, and Filipa Sa. "The U.S. Current Account and the Dollar." National Bureau of Economic Research Working Paper Series No. 11137 (2005).

Borensztein, Eduardo, Kevin Cowan, Barry Eichengreen, and Ugo Panizza, eds. Building Bond Markets in Latin America: On the Verge of a Big Bang? Cambridge, MA: MIT Press, 2008.

Burger, John D., Francis E. Warnock, and Veronica Cacdac Warnock. "Investing in Local Currency Bond Markets." National Bureau of Economic Research Working Paper Series No. 16249 (2010).

Caballero, Ricardo J. "On the Macroeconomics of Asset Shortages." National Bureau of Economic Research Working Paper Series No. 12753 (2006).

———. "The "Other" Imbalance and the Financial Crisis." National Bureau of Economic Research Working Paper Series No. 15636 (2010).

Chinn, Menzie D. "Getting Serious About the Twin Deficits." In The Bernard and Irene Schwartz Series on the Future of American Competitiveness. New York: Council on Foreign Relations, 2005.

Chinn, Menzie D., and Hiro Ito. "Current Account Balances, Financial Development and Institutions: Assaying the World "Saving Glut"." Journal of International Money and Finance 26, no. 4 (2007): 546-69.

Clarida, Richard H. 2005. Japan, China, and the U.S. Current Account Deficit. Cato

Journal 25 (1).

Clement, Douglas. "Interview with Ricardo Caballero." The Region, June 2011 2011, 28-31.

Cordeiro, Anjali. "Emerging-Market Debt Records Strong Year." Wall Street Journal (Online), December 31 2010.

Eichengreen, Barry. "The Dollar Dilemma: The World's Top Currency Faces Competition." Foreign Affairs 88, no. 5 (2009): 53-68.

Eichengreen, Barry, Menzie D. Chinn, and Hiro Ito. "A Forensic Analysis of Global Imbalances." La Follette School Working Paper No. 2011-007  (2011).

Erceg, Christopher J., Luca Guerrieri, and Christopher Gust. "Expansionary Fiscal Shocks and the Us Trade Deficit*." International Finance 8, no. 3 (2005): 363-97.

Flood, Chris. "State Street Brings New Etfs to London." Financial Times (2011), http://www.ft.com/intl/cms/s/0/50e4b510-b6d0-11e0-a8b8-00144feabdc0.html - ixzz1TKOfXpuN.

———. "Van Eck Launches Local Currency Em Bond Etf." Financial Times (2010), http://www.ft.com/intl/cms/s/0/317f757a-9679-11df-9caa-00144feab49a.html - axzz1TK1GXVKA.

Forbes, Kristin J. "The Microeconomic Evidence on Capital Controls: No Free Lunch." In Capital Controls and Capital Flows in Emerging Economies: Policies, Practices, and Consequences, edited by Sebastian Edwards. Chicago: University of Chicago Press, 2007.

Gavin, Michael, and Jose Wynne. "Em Strategy: Local Versus External under a 'New Norm'." In Barclays Capital Emerging Market Research: Barclays Capital, 2010.

"The Globalization of Corporate Finance in Developing Countries." In Global Development Finance 2007. Washington, DC: World Bank, 2007.

Goodman, J., and L. Pauly. “The Obscolescence of Capital Controls? Economic

Management in an Age of Global Markets.” World Politics 46, 1993: p. 50-82

Greenspan, Alan. "Current Account: A Speech at the Advancing Enterprise 2005 Conference." London, England, 2005.

———. "Mortgage Banking: A Speech to the American Bankers Association Annual Convention (Via Satellite)." Palm Desert, CA, 2005.

Gruber, Joseph W., and Steven B. Kamin. "Explaining the Global Pattern of Current Account Imbalances." Journal of International Money and Finance 26, no. 4 (2007): 500-22.

Guha, Krishna. "Paulson Says Crisis Sown by Imbalance." Financial Times, January 1 2009.

IMF, “Global Liquidity Expansion: Effects on “Receiving‟ Economies and Policy

Response Options,” Global Financial Stability Report, April 2010

Ito, Hiro, and Menzie Chinn. "East Asia and Global Imbalances: Saving, Investment, and Financial Development." National Bureau of Economic Research Working Paper Series No. 13364, no. published as Hiro Ito, Menzie Chinn. "East Asia and Global Imbalances: Saving, Investment, and Financial Development," in Takatoshi Ito and Andrew K. Rose, editors, "Financial Sector Development in the Pacific Rim, East Asia Seminar on Economics, Volume 18" University of Chicago Press (2009) (2007).

Lin, Cynthia. "Debt Issues in Emerging Markets Has a Draw." Wall Street Journal (Online), 2011.

Lucas, Robert. "Why Doesn't Capital Flow from Rich to Poor Countries?" American Economic Review 80, no. 2 (1990): 92-96.

Márcio Gomes Pinto, Garcia, and S. de M. Carvalho Bernando. "Ineffective Controls on Capital Inflows under Sophisticated Financial Markets: Brazil in the Nineties." Department of Economics PUC-Rio (Brazil), 2006.

"Market Vectors Etfs: Investment Case for Emerging Markets Local Currency Bonds." New York: Van Eck Global, 2010.

Mendoza, Enrique G., Vincenzo Quadrini, and Jose-Victor Rios-Rull. "Financial Integration, Financial Deepness and Global Imbalances." National Bureau of Economic Research Working Paper Series No. 12909 (2007).

"Narrowing the Gap - a Clarification of Moody's Approach to Local Vs. Foreign Currency Bond Ratings." In Sovereign Methodology Update: Moody's Investors Service, 2010.

Obstfeld, Maurice. "International Finance and Growth in Developing Countries: What Have We Learned?" National Bureau of Economic Research Working Paper Series No. 14691 (2009).

Obstfeld, Maurice, and Kenneth Rogoff. "Global Imbalances and the Financial Crisis: Products of Common Causes." In Federal Reserve Bank of San Francisco Asia Economic Policy Conference. Santa Barbara, CA, 2009.

Peiris, Shanaka. "Foreign Participation in Emerging Markets  Local Currency Bond Markets." In IMF Working Paper Series. Washington, DC: International Monetary Fund, 2010.

"Rebalancing Growth." In World Economic Outlook. Washington, DC: International Monetary Fund, 2010.

Rogoff, Kenneth. "Global Imbalances and Exchange Rate Adjustment." Journal of Policy Modeling 29, no. 5 (2007): 705-09.

"Statement of G7 Finance Ministers and Central Bank Governors, August 8." 2011.

Summers, Lawrence H. "Reflections on Global Account Imbalances and Emerging Markets Reserve Accumulation. Draft as Prepared for Delivery, L.K. Jha Memorial Lecture, Reserve Bank of India." Mumbai, India, 2006.

Turner, Philip. "Secondary Market Liquidity in Domestic Debt Markets." In Paper Prepared for the Tench Annual OECD/World Bank/IMF Bond Market Forum, April 24, 2008.

Warnock, Francis E., and Veronica Cacdac Warnock. "International Capital Flows and U.S. Interest Rates." National Bureau of Economic Research Working Paper Series No. 12560 (2006).

Wheatley, Jonathan. "Fund Flows: Em Bonds Gain over Equities Despite Greece." Financial Times, June 17, 2011 2011.

Wigglesworth, Robin. "Rouble Debt Hits the Spot for Investors." Financial Times, July 4 2011.

Wigglesworth, Robin, and Stefan Wagstyl. "Search for Yield Boosts Emerging Economy Debt." Financial Times, May 23 2011.