Abstract
Heavily reliant on the work of Charles P Kindleberger, theories of hegemony and power transition have been built on the largely unchallenged assumption that the United States acts as a global economic stabilizer in time of crisis. However, there has been no attempt systematically to delineate, operationalize and test whether the United States performs this role utilizing Kindleberger’s five functions. Proponents of these theories have, in contrast, characterized China as either a free-rider or a predator waiting to challenge American leadership in times of crisis. In evaluating these characterizations, we test two hypotheses that examine the extent to which the United States and China have performed these stabilizing functions with regard to three major economic crises (the 1997 Asian economic crisis, the 2001 “DotCom” crisis and the Great Recession of 2007–2009). Our findings suggest that Kindleberger’s functions are more evenly shared than conventional scholarship would predict and that China has played a supportive, stabilizing role. Without evidence of explicit collaboration between the two countries, these functions were shared. Furthermore, China played an increasingly important role in supporting the global economic system over time and the trend line suggests they will continue to do so in the future. If Kindleberger’s criteria are correct, the assumption of a single country, the United States, acting as a stabilizer is therefore empirically mistaken. If his criteria are flawed, major IR theories that have assumed the criteria to be true and/or have assumed that the United States plays this unique role in the global economy are in need of critical reassessment.
The role of the lender of last resort is fraught with ambiguity and dilemma. There are times when rules and precedents cannot be broken. Others when they cannot be adhered to with safety.
Introduction
More than seven years after the Great Recession began, there is only now emerging a substantial body of work in International Relations (IR) and International Political Economy (IPE) dedicated to what the literature has characterized as one of the greatest challenges to the functioning of global capitalism in almost a century (Drezner, 2014; Helleiner, 2014; Helleiner and Pagliari, 2011; Kirshner, 2014; Schwartz, 2010). As Joseph Stiglitz noted, tens of millions lost their jobs (over 20 million in China alone) and more fell into poverty (Stiglitz, 2010). Investors lost trillions of dollars, stock markets plunged and the global financial system threatened to become illiquid.
In this article we seek to contribute to this literature by addressing two questions relating to crisis management, one theoretical and the other empirical. Our theoretical question concerns whether a single leader is necessary for economic stabilization in such a crisis. This question remains relevant. Hegemonic stability theory was discredited in the 1980s (cf. Keohane, 1984; Snidal, 1985); however, the assumption that a single leader is required to lead in the stabilization of the international economy persists in the IR and IPE literature. It is used to justify a variety of contemporary theories from contrasting perspectives – including those relating to both transition of power and economic crisis management (Drezner, 2014; Ikenberry, 2011; Tammen et al., 2000). Liberals, for example, have argued “that America or the global institutions it created have continued” to play a dominant, if not unique role in these crises (Drezner, 2014; Ikenberry, 2011). Power transition theory as articulated by Offensive Realists, in contrast, would lead us to presuppose that China, as a potential rival to US leadership, might use such crises as an opportunity to challenge the prevailing rules of the global system and thereby undermine American leadership (Mearsheimer, 2001; Tammen and Kugler, 2006; Tammen et al., 2000).
The second related empirical question concerns “who did what” during the Great Recession? Did the United States demonstrate leadership or were these functions shared in the management of the crisis? We examine these questions through the lens of the work of Charles Kindleberger, which both Liberals and Realists regard as foundational when examining the dynamics of crisis management. For example, Kindleberger’s 1973 book, The World in Depression, has been cited over 2560 times according to Google Scholar (Kindleberger, 1973). However, despite Kindleberger’s profound impact, neither Liberals nor Realists have tested Kindleberger’s argument in its original form, using all the functions and relevant indicators he specified. Thus, paradoxically, while scholarly work has moved far away from his original formulation, Kindleberger’s insights regarding the causes of the Great Depression remain central to a shared understanding about how the global economic system functions and (prescriptively) should best function.
Our underlying objective in this article is therefore to examine four issues. First, to consider the continued utility of Kindleberger’s work; second, to evaluate the continued centrality of the USA to the global economy in crisis; third, to examine the form and degree of China’s emergent role as a stabilizer (Johnston, 2003; Kang, 2007); and finally to assess empirically whether stabilization functions can be shared in a crisis (Keohane, 1982, 1984; Oye, 1986; Zürn, 1992).
To do so we therefore return to Kindleberger’s original formulation, focusing on the economic indicators he specified. As we demonstrate, his work is often used without consideration of the circumstances to which he applied his claims – in what he characterized as “panics” (Kindleberger, 1978: 4–5). Our article is therefore fundamentalist in character: it attempts to employ Kindleberger in terms of his expressed original intent and form, in examining the question of stabilization in crisis.
In doing so, we consider two specific hypotheses. The first tests Kindleberger’s assertion that “for the world economy to be stabilized, there has to be a stabilizer – one stabilizer” (Kindleberger, 1973: 304). The second tests whether China shared the role of crisis stabilizer with the United States (US). We compare US and Chinese behavior in terms of Kindleberger’s functions, examining data from three recent crises: the Asian crisis of 1997–1998, the “DotCom” bubble crisis of 1999–2001, and the Great Recession of 2007–2009.
However, in doing so we acknowledge the limitations of our tests. First, these three crises are not directly comparable because of their varying scope. We therefore use the first two crises as a benchmark, in order both to assess trends across time and to highlight the greater severity, magnitude and significance of the Great Recession. Second, we recognize that China is disadvantaged in this comparison because the US provides the primary international currency and because American financial markets are so much more developed. However, as the results reveal, China has played an increasingly influential role despite this constraint. Third, while Kindleberger argued that the magnitude of state intervention would vary with the magnitude of crisis, he rejected the possibility that different crises might pose different challenges for nation states and would thus require a different policy response (Kindleberger, 1978: 190–200). Ideally we would include a measure of the consequences of various policies, i.e., when and to what extent remedies have a discernible effect. But because Kindleberger believed that performing the five functions was synonymous with stabilizing the international economy in the face of crisis, he does not provide us with sufficient material to test such boundary conditions. Whereas Kindleberger’s neglect of state policy as a determining factor might be a legitimate critique of his approach (Drezner, 2014), it is clearly not one we ought to incorporate when testing “his” theory.
If Kindleberger is correct, that there has to be one stabilizer (in practice the United States) on all of the five functions he specified during crises – as measured by our indicators. For confirmation of the second hypothesis – that China played a role as a stabilizer – we should expect it to provide at least one of these functions to a greater extent than the United States.
We exclude Europe in our comparison for two reasons. First, preliminary evidence drawn from the Great Recession suggests that the European Union –beset by financial crisis among several members – was unable to provide the necessary resources to contribute as a stabilizer on a global scale. Indeed, European governments had to rely extensively on Chinese purchases of Euro-bonds to address its short-term financial needs (Reich and Lebow, 2014: 112–114). Second, Keynesian-style expansionist policies are a requisite for performing stabilization functions. Although current account data demonstrates that Germany nominally performed this regional role for these European peripheral countries, other major European countries favored austerity policies (the UK) or lacked the requisite resources (France) to act (Blyth, 2013; Schwartz, 2014). Only the US and China, the two largest economies, had the capacity to act as a stabilizer in crisis during the Great Recession. 1 Kindleberger’s formative assessment focused on a two-country (US–UK) comparison (Kindleberger, 1973). We therefore do likewise here.
The remainder of the article is divided into four sections. In the next section, we outline briefly the way that Kindleberger’s work has been used. We then offer a schema for testing Kindleberger’s hitherto untested proposition. In the third section we then apply it to the study of both American and Chinese behavior across 16 years and three crises. The article’s concluding section reflects on the possible implications of our findings for current popular theories of both leadership and crisis management, and the pattern it may portend for the future functioning of global economic management.
Kindleberger in the IR and IPE literature
Kindleberger’s original formulation has been adapted, amended and utilized in varying ways. We can only illustrate briefly the extent to which it has remained a consistent reference point over the last four decades.
Amongst Realists, for example, Robert Gilpin notes,
Stephen Krasner and I each appropriated Kindleberger’s basic idea that a political leader was needed to create and manage an international liberal economy. However, each of us made several modifications that placed Kindleberger’s insight within a state-centric intellectual framework of political analysis and thus fashioned a state-centric version of the theory of hegemonic stability. (Gilpin, 2001: 99)
Stephen Krasner linked political centralization to the openness of international trade by focusing on tariff levels, trade proportions and regional trading patterns as indicators (Krasner, 1976: 324–330). He cited Kindleberger more overtly as influential in linking power capabilities to international stability and openness in a subsequent article with Michael Webb (Webb and Krasner, 1989). Krasner and Webb performed an empirical test of their proposition; however, it is not confined to periods of crisis, and their functions did not match those specified by Kindleberger. 2
In Robert Keohane’s seminal work, After Hegemony, he characterized these Realist formulations as “basic-force models” in which the relationship between power, interest and (in effect) stabilizing leadership is automatic and thus unproblematic (Keohane, 1982: 32–34). In them, a capacity and willingness to act as a leader are organically related in producing international order. Keohane stressed the tautological nature of this formulation, in that a leader both capable and willing to underwrite a stable international order will – by definition – provide it, resulting in a non-falsifiable proposition (Keohane, 1982). He notes that it remains interesting to examine the capability and willingness to lead in order to describe the international system (Keohane, 1982: 35). Consequently, while Kindleberger’s work has been used as a way to explain the stability of international order as a function of a dominant actor’s ability and willingness to perform the five functions, we concur with Keohane that there is empirical utility in evaluating “who does what” in behavioral terms.
A plethora of American IPE subsequently developed on the basis of the Liberal Institutionalist and Realist positions. Joanna Gowa examined trade barriers as an indicator of hegemony (not of crisis management), adding security considerations. She concluded that “casual empiricism” seemed to confirm Kindleberger’s argument (Gowa, 1989: 310). Mark Brawley commented that, “[i]n as much as the Amsterdam market was the international market, the Bank of Amsterdam provided some of the market stabilizing functions Charles Kindleberger has argued a hegemon must perform” (Brawley, 1993). However, Brawley did not examine Kindleberger’s functions systematically. Helen Milner (1988: 355) criticized Kindleberger’s idea that a lag accounts for differences in hegemonic trade policies. According to Beth Simmons (1994), Kindleberger’s argument about economic leadership is more conditional than portrayed in the literature that cites him. She stresses that Kindleberger believed that the hegemon was as much motivated by domestic politics as external incentives (Simmons, 1994: 5–6). In effect, domestic politics considerations may partially or entirely prevent the hegemonic power from perform stabilization functions or require it to deliberately or inadvertently share these roles (Simmons, 1994: 5–6). In Liberal Leviathan, G. John Ikenberry (2011) assumed the veracity of Kindleberger’s position in commenting that the United States provided public goods during the Cold War. As in all these examples, scholars provided no systematic examination of Kindleberger’s functions consistent with his formulation to reinforce this supposition. In each case, the discussion shifted from stabilization in crisis to routine American leadership. Furthermore, these authors invoked Kindleberger’s formulation to develop the concept of hegemony, even though he explicitly disavowed it in the pages of International Organization (Kindleberger, 1986).
Two recent studies, by Benjamin Cohen (2012) and Daniel Drezner (2014: 123–164), place far greater emphasis on Kindleberger’s functions. Both focus on the dimensions of economic leadership he explored and contribute to our understanding of the dynamics of the Great Recession and the post-2008 global economy. Cohen explicitly employs Kindleberger’s five functions, although his concern is whether China’s rise as a monetary power can be successfully accommodated, rather than China’s explicit role as a stabilizer in the Great Recession (Cohen, 2012). Drezner (2014) does consider the sources of what he characterizes as the resilience of the global economy in the Great Recession. He draws selectively from Kindleberger’s criteria, supplementing it with other functions outside the scope of Kindleberger’s formulation. He argues that:
Despite initial shocks that were more severe than the 1929 financial crisis, global economic governance responded quickly and robustly. Whether one looks at economic outcomes, policy outputs, or institutional resilience, these governance structures either reinforced or improved upon the status quo after the collapse of the subprime mortgage bubble. These regimes performed particularly well during the acute phase of the crisis in the fall of 2008, ensuring the continuation of an open global economy… Simply put, the system worked – the open global economy survived because of ‘good enough’ global governance. (Drezner, 2014: 124)
However, Drezner (2014: 153) is more ambivalent having examined the evidence, noting that “American power and leadership during the recent crisis turned out to be more robust than many experts perceived.” He concludes, somewhat puzzlingly, that, “[t]he evidence suggests that multilateral institutions adapted and responded to the 2008 financial crisis in a robust fashion. They passed the stress test – global economic governance has been good enough” (Drezner, 2014: 156). Drezner’s argument can therefore be interpreted in one of two ways: either global institutions or the United States played the role of a stabilizer. In neither case does he consider the proposition that critical stabilization functions were shared at the national level between the United States and China, having little to do with the resilience of global institutions.
IR and IPE scholars often still cast the US as a continued leader in the global economy, despite the fact that Kindleberger himself emphatically stated that the US had abandoned that role several decades ago (Kindleberger, 1981a). Our logic in responding to this scholarship is simple: if scholars rely so heavily on Kindleberger’s criteria about what an economic stabilizer should do, and they unquestionably assume that the US performs these tasks, in building theories of leadership, because of the centrality of the US economy to the functioning of the global system, then we should investigate whether they are right by testing using the criteria he proposed.
Formulating a test of the role of China and the US in three economic crises
Kindleberger collectively listed five key stabilizing functions: maintaining an open market for distress goods; providing countercyclical lending; policing a stable system of exchange rates: ensuring coordination of macroeconomic policies; and acting as a lender of last resort (Kindleberger, 1973: 305; 1981a: 247; 1981b). 3 However, Kindleberger himself eschewed testing his assertion using aggregate statistical data. Rather, he assessed his argument in a broader historical study, Manias, Crashes and Panics – where he defined crises in terms of the collapse of speculative bubbles – and in a subsequent collection of essays (Kindleberger, 1978, 1981b). 4 Nonetheless, any discussion of the stabilizer’s five functions remains limited (Kindleberger, 1973: 305). As a result, one of these functions – the role of a “lender of last resort” – has itself become a shorthand term among contemporary scholars, bankers and policymakers for America’s purported postwar role as underwriting the stable functioning of the global system. 5
We delineate, codify and operationalize the five functions specified in Kindleberger’s original formulation. Like Kindleberger, we distinguish between routine leadership and a stabilizing role in times of crisis. We then analyze successive American behavior across three crises, culminating in the Great Recession, and compare it to China’s behavior as the other member of what Geoffrey Garrett (2010) labels the “G2 in G20.” Doing so has the advantage of allowing us to consider variant behavior given the differing magnitude and scope of each crisis: 1997–1998 was a regional one in Asia, 1999–2001 was centered in the US technology sector, and 2007–09 was on a global scale. 6
The scope of our analysis is inevitably limited by the availability of data. Our time series analysis covers 18 years – from 1996 to 2014. We would have preferred to examine the comparative behavior of these two countries over a longer period of time, but much of the reliable data we would require to do so for China does not predate 1996. Nonetheless, we believe that evidence spanning nearly two decades and three crises does provide a reasonable indication of American and Chinese behavior – both during crises and in more stable periods. It also helps us analyze possible trends for the future.
To satisfy the distinct requisites of both Kindleberger and those who have embraced but adapted his work, we divide the analysis of our findings into three components. The first is the behavior of both countries during routine times. This spans several periods: from early 1996 to mid-1997; the end of 1998 to mid 1999; early 2001 to early 2007; and early 2009 until the date of the most recently available data. The second is each country’s behavior during a crisis – the Asian financial crisis 1997–1999, the DotCom crisis from late 1999 to early 2001, and the Great Recession early 2007 to early 2009. Our third and final component concerns the trend of each country over time and thus provides for extrapolation into the future.
In order to justify our claim that only the Great Recession constituted a global crisis, we distinguish and measure the magnitude of financial crises from more routine periods of stability and growth. To do so, we simply trace the volatility in three major equity markets: the Dow Jones World Index, the Index for South East Asia and the Nasdaq Index, each measured in terms of the annualized standard deviation of returns. 7 We recognize that an escalation in equity prices and their attendant collapse is only one indicator of financial crisis, although economists do regard it as a leading indicator (Reinhart and Rogoff, 2008). Moreover, not all financial crises expand into economic crises, resulting in economic recession or depression. However, as Figure 1 makes clear, the magnitude of volatility is inordinately greater than the two prior crises. We therefore recognize that the three crises under consideration are not directly comparable in terms of scope and magnitude. However, the comparison of the three has utility because our aim is to demonstrate that the 2008 crisis is closest in order of magnitude to the Great Depression examined by Kindleberger.
Financial crises in the period 1996–2013.
In Figure 1, the vertical grey lines isolate the Asian financial crisis when volatility increased in South East Asia (Dow Jones South East Asia Index); the subsequent dotted black lines demarcate the dotcom crisis when volatility increased in the technological sector (Nasdaq index); and, finally, the black lines delineate the Great Recession when volatility increased across all three indexes (the Dow Jones World Index and the other two indexes). The data illustrate that the Great Recession is the only crisis when volatility was high and moved in tandem across all three indices, indicative of the global scale of that crisis.
We divide our analysis of the findings into three categories: leadership over the balance of the period from 1996–2013; stabilization behavior during the crises; and the future trend based on current trajectories. Before presenting our empirical results and analysis, however, we highlight important methodological issues in terms of selecting and justifying the indicators for the various functions.
Operationalizing Kindleberger’s five functions
As our brief literature survey demonstrates, we acknowledge and accept that all of the scholars cited above have argued that not all of these five functions are relevant for the stabilization or leadership of today’s global economy, even as they paradoxically relied on his work in analyzing American leadership. Many have focused instead on government policy (cf. Drezner, 2014). Clearly, we differ from their view because we examine the behavior of nation states in times of a global crisis in the twenty-first century – just as Kindleberger did in the twentieth. In future research we intend to examine how functions vary between both crisis and routine periods. Such an investigation, however, is beyond the scope of this article. The definition and technical measure of each indicator is outlined below, with each subsequently illustrated by appropriate data.
A market for distressed goods. The first function in Kindleberger’s formulation is the provision of a market for distressed goods. It is relatively straightforward to measure in terms of world import shares and imports as a share of GDP. Kindleberger is very clear that the country providing this function must be able to “…adapt domestic resources to changes in productive capacities abroad and…to keep the import market open in periods of stress. The first is more readily done by a rapidly growing country…” (Kindleberger, 1973: 291). Using additional measures to gauge the extent to which countries adapt to growing import-competition is fraught with difficulties. For instance, there are difficulties in isolating the employment rate in the import-competing sector and it is hard to determine whether changes in the employment rate in the import-competing sector are due to import-competition or fluctuations in domestic demand. We restrict our indicator to world import shares because only larger economies have a sufficiently sized market to absorb other countries’ exports consistent with being a haven for distressed goods. Because large, fast-growing countries should be more willing to accept the reallocation of resources and the restructuring of the economy implied by import-competition relative to big slow-growing ones, this measure also captures the capacity for adaptation mentioned by Kindleberger. Several complementary indicators could be used to measure the extent to which countries keep import markets open in periods of stress such as tariffs and non-tariff barriers (NTBs) as well as currency manipulation. However, we focus on world import shares to evaluate how much trade a country engages in relative to the rest of the world and its share of imports with regard to GDP to evaluate how much trade a country absorbs relative to the size of its economy. While there are market-based reasons why a country’s share of imports fluctuates, whether as a share of the world total or relative to GDP, they are a direct and more reliable estimate of foreign import penetration. 8
The provision of a counter-cyclical flow of capital is measured using each country’s share of world capital exports. Kindleberger did not distinguish between different forms of capital (short- or long-term capital, or between portfolio and direct investment) although the very notion of counter-cyclical capital suggests more liquid forms of capital.
The provision of stable exchange rates. To measure this indicator, we consider volatility – measured as the annualized standard deviation of the change – in nominal and effective exchange rates. 9 Kindleberger advocated stable (even fixed) exchange rates, which he saw as facilitating cross-border trade and investment by increasing the predictability of cross-border flows. Scholars, however, do not see exchange rate stability as a panacea (cf. Deudney and Ikenberry, 2012: 1). Flexible exchange rates can promote stability by adjusting macroeconomic, trade and financial imbalances. 10
The coordination of macroeconomic policies. Kindleberger specifically called for the systemic leader to “provide a degree of coordination of domestic monetary policies” (Kindleberger, 1981a: 247). This indicator evaluates whether governments and central banks are deliberately acting in concert. Through policy statements, we examine whether countries are following the US or China through joint monetary easing or tightening.
Acting as lender of last resort. Benjamin Cohen notes that, “…[b]oth ample reserves and an internationalized currency enable a country to act, if it wishes, as a lender of last resort in times of crisis—a source of liquidity for others” (Cohen, 2012: 3). We compare liquidity creation by the United States and China because only the expansion of the Central Bank’s monetary base (not the buildup of reserves) creates liquidity. We also include China’s reserves because China’s reserve accumulation exchanges liquid capital for reserve assets (mostly dollar denominated reserves). We do not include American reserves since the dollar is the primary international currency and therefore has minimal reserves that would barely be visible in any graph. We also use refinancing rates, those at which banks lend to each other in the interbank market, and swap arrangements for foreign currency funding in times of crisis. In terms of refinancing rates, we compare the interbank cost of borrowing funds in dollars and renminbi using 3-month US dollar LIBOR rates and 3-month SHIBOR and 3-month CHIBOR rates. The 3-month CHIBOR was introduced in 1996 and is based on trades whereas SHIBOR, introduced in 2007, is based on price quotes and therefore seen as a better more market-sensitive benchmark. (Conway, et al., 2010: 9).
Empirical findings
To what extent did the US and China each perform these five functions? Except for Kindleberger’s fourth criterion (coordination of macroeconomic policies, which we discussed in detail), each function is represented below by graphs comparing the relative performance of each country with respect to the indicator used in order to quantify the function in question. Each indicator is examined for the periods between 1996 and the latest date for which there are data. Where appropriate, we include a trend line. 11 We place special emphasis on the period within which the Great Recession unfolded (between February 2007 and March 2009, based on the timeline established by the US Federal Reserve) because Kindleberger was specifically interested in the role of an economic leader during times of global crisis. Since he did not attempt to weigh the relative importance of the various functions, we do not do so either.
The market for distressed goods
The US has been the world’s largest importer in the postwar period. During the Great Recession, America’s share of world imports declined slightly, from 14% to 13%, whereas China’s initially declined, from 8% to 6%, but then increased to 11% percent. In 1996, the US accounted for 15% of world commercial imports whereas China absorbed 3%. By 2011, the US share declined to 12% whereas in contrast China’s share increased to 10%; the gap between the two countries had shrunk to just 2 percentage points. As Figure 2 illustrates, if this trend continues, China will overtake the US as the world’s largest importer within a decade.
(1) Market for distressed goods: Share of world imports.
According to an alternative measure, China has been more open to trade than the US over the entire period (see Figure 3). In routine periods, China has had a larger share of imports relative to the size of its economy than the United States. During the Great Recession, however, the Chinese import share dropped more sharply than did the American. However, both the American and Chinese share of imports with regard to GDP has risen over time. Despite some convergence, the trend suggests that the Chinese share of imports will remain higher than America’s for the foreseeable future.
(2) Market for distressed goods: Import share with respect to GDP.
In sum, the US remains the world’s first market for distressed goods if imports are measured as a share of the world’s total, whereas China is more open to foreign goods relative to the size of its economy. During the Great Recession, American imports declined more than China’s relative to world imports, while Chinese imports declined more substantially relative to GDP.
The trend line suggests that China is set to surpass the US as the world’s largest importer and is also likely to continue importing more goods relative to the size of its economy. This trade pattern persisted through the first two crises. However, the composition of the two countries’ imports differs markedly: China is primarily an important market for commodity producers. Although Kindleberger is vague in his definition over what constitutes distressed goods, both raw materials and energy do not fit easily in this category since they can be traded elsewhere and sometimes be used domestically to greater effect as in the case of Brazil’s rare earth metals or Africa’s oil reserves, wood and minerals. Not much will change in terms of commercial leadership if China displaces the US as the world’s largest importer without moving up the value-added ladder, which it has done to some extent, though not comprehensively. Only countries capable of running serial deficits and absorbing the international supply of goods across a range of goods can sustain commercial leadership (Germain and Schwartz, 2014; Norrlof, 2009, 2010). Nevertheless, if Chinese firms consume more sophisticated products, countries will increasingly view China as the main locomotive in the world economy.
Counter-cyclical flows of capital
The US has been the world’s largest exporter of capital during normal times and acted that way during the first two crises. Yet its leading role diminished as the Great Recession unfolded. As Figure 4 illustrates, the US repatriated more capital than it exported and China overtook the US as the world’s primary capital exporter in the first half of the crisis. China’s increasing share of capital exports has primarily taken the form of a buildup of official reserves and accelerated foreign direct investment and portfolio investment through vehicles such as sovereign wealth funds (Reich and Lebow, 2014: 121–125). However, these flows remain nominal – in comparison to American flows and as a proportion of China’s reserves. Given that China’s advance over the US in capital exports is one-dimensional, it is difficult to state definitively whether this is stabilizing or not. A case can be made that China has played a stabilizing role by exporting capital in order to accumulate US dollar and Euro reserves through purchases of American and European government debt. The large global imbalances that were seen as a threat to the global economy could be successfully accommodated because China continues to export capital to the United States. Consequently, the decline in the value of the US dollar and the sharp increase in interest rates that some scholars anticipated failed to materialize. Conversely, China’s export surplus with the US, made possible by exporting capital to the US, was one of the causes of what scholars have identified as a key determinant in destabilizing global imbalances (Aizenman, 2010; Freund, 2010). According to this interpretation, China’s accumulation of dollar reserves was disruptive, principally because it failed to diffuse capital exports more broadly in the international economy as a stabilizer is expected to do. Although America’s cross-border supply increased above China’s after the financial crisis, the trend lines suggest China’s supply of capital will exceed America’s supply in the near future.
Counter-cyclical flow of capital: China’s capital exports relative to the United States.
In sum, the US is the world’s leading provider of liquidity, although China did provide some liquidity during the Great Recession by trading cash for liquid reserve (mostly US dollar) assets. The long-term trend is for the US to remain the primary liquidity creator and provider. If anything, the Euro is currently a more potent rival to the dollar, although it suffers from a series of deficiencies that make it likely to unseat the dollar anytime soon (Feldstein, 1997; Cohen, 2003; Posen, 2008; Norrlof, 2009, 2014).
Stability of exchange rates
Figure 5 tracks the volatility of the respective currencies relative to a trade-weighted basket of currencies. In routine periods, the dollar’s nominal exchange rate has been more volatile than the renminbi’s (RMB). This pattern was maintained during the first two crises. The Great Recession was also marked by greater instability in the nominal exchange rate of the dollar than that of the RMB. Although variability in the dollar rate is declining, the trend lines suggest the dollar is likely to remain more volatile than the RMB.
(3.A) Stability of exchange rates: Volatility in nominal effective exchange rates.
In contrast, as Figure 6 demonstrates, no country clearly leads in stabilizing exchange rates when the dollar’s and RMB’s value relative to other currencies is adjusted for inflation. In normal times, the RMB is periodically more volatile than the dollar in some periods, whereas the converse is true during others. Similarly, neither country was clearly a stabilizer during the first two crises nor during the Great Recession—the dollar is less volatile in the first period but not the second. However, the trend suggests the dollar’s real exchange rate is becoming more volatile than the RMB over the long term.
(3.B) Stability of exchange rates: Volatility in real effective exchange rates.
A major reason for the greater stability in the renminbi exchange rate than in the dollar’s is the latter’s international role. The US dollar is the world’s most traded currency. It is used most widely to denominate assets and goods as well as being the currency figuring most prominently in government reserves. The renminbi’s international role, while growing, is relatively insignificant. Paradoxically, China cannot assume a greater role in financial and commercial markets unless the renminbi is traded more freely and used more extensively for the settlement of trade and financial transactions. The new Chinese government has made incremental efforts to do so. If the value of the renminbi is to a greater extent determined in foreign exchange markets, we should also expect more volatility in the renminbi exchange rate. So while the renminbi has been more stable than the dollar during the period we examine, questions remains as to whether such stability is compatible with its growing international role and how Chinese governments will address this developing paradox.
Macroeconomic policy coordination
Macroeconomic policy coordination includes efforts at joint monetary expansion (or tightening). We explore here whether such moves were mostly organized or triggered by the US or China. Actions taken under the auspices of the International Monetary Fund (IMF) are excluded from this analysis because we want to establish if there is a clear “follow-the-leader” effect. In this regard, internal bargaining in international institutions is a confounding variable (Putnam, 1988). Thus our findings potentially understate the US’ impact on concerted efforts to stabilize the economy given its prominent voting rights in the organization.
Towards the end of the 1990s, the US and Japan intervened jointly to prevent the yen’s further depreciation. Subsequent efforts were limited until the US Federal Reserve temporarily established thirty-day dollar swap lines in the immediate aftermath of the 11 September 2001 terrorist attacks on the United States. China played a limited role in both of the first two cases.
A period of relative inactivity followed until the onset of the Great Recession, when a series of major organized liquidity provision actions were implemented through a process of monetary easing. In October of 2008, central banks in the US, UK, Canada, Sweden, and Switzerland lowered interest rates by half a percentage point, while China’s central bank separately cut its key rate by just over a quarter of a point (Lanman, 2008). The Board of Governors of the US Federal Reserve also lowered the discount rate by 50 basis points to 1.75% (BGFRS, 2008). Both Hong Kong and Taiwan followed suit in lowering interest rates.
To support the liquidity injection, in November of 2008 the Federal Reserve cut the rate foreign central banks pay to borrow US dollars by 50 basis points (to 0.5%) in a coordinated move with the European Central Bank and central banks in Britain, Japan, Switzerland, and Canada. Within days, China lowered reserve requirements for financial institutions by 50 basis points.
In sum, there is some evidence of US-initiated multilateral macroeconomic policy coordination during the Great Recession; activity which, however, did not involve the Chinese. When attempts have been made to harmonize policies, a clear pattern of follow-the-leader emerges with the US the first to ease monetary policies and others following suit.
Lender of last resort functions
An important aspect of acting as lender of last resort is the supply of an internationalized currency. Kindleberger is less specific in defining it than is the case with other functions. This problem is complicated by the fact that a comprehensive comparison of the relative role of the dollar to the renminbi is not feasible because of the lack of data regarding the use of the renminbi. We therefore offer three indicators for comparing the behavior of the two countries.
First, the US dollar remains unquestionably paramount, with dollars used by both official and private investors to a far greater extent than any other currency in the world. Figure 7 provides a snapshot of the relative use of the dollar and the RMB in global foreign exchange markets, which is the best proxy for overall use given the dearth of RMB data.
(5.A) Lender of last resort functions: Global currency role of USD and RMB.
As Figure 7 demonstrates, the dollar was used far more widely than the RMB throughout the period under investigation – both in normal times and during all three crises. According to the trend lines, the dollar will continue to be used to a much greater extent than the RMB, for the foreseeable future (Cohen and Benney, 2013; Norrlof, 2014).
Given the scant usage of the renminbi, Figure 8 offers a second lens through which to consider lender of last resort functions. It compares American and Chinese lender of last resort tasks in terms of liquidity provision as measured through the United States’ supply of dollars and Chinese official reserve holdings.
(5.B) Lender of last resort functions: Liquidity.
In terms of liquidity creation, the US was the leading liquidity provider during all three crises: America’s monetary base (the thick black line in Figure 8) is located well above China’s monetary base (the dotted line). According to this measure, the United States provided more liquidity than China during the Great Recession. The long-term trend is for the United States to continue to be the leading source of liquidity in the global economy. As is also evident in Figure 8, China provided liquid capital in exchange for reserve assets (the dashed black line). China did contribute therefore to crisis stabilization during the Great Recession, by not engaging in panic selling.
A third and final lens examines the ability to affect the inter-bank market where banks lend to each other. Both the Federal Reserve and the People’s Bank of China have the capacity to do so and Figure 9 contrasts interbank lending in the respective currencies. The US is a clear leader in this domain, offering lower rates on dollar-based loans during normal times; Chinese RMB rates were lower for a very brief period (mid-1999 to mid-2001) i.e., during the DotCom crisis. The available data illustrate that throughout the Great Recession the US took on the role of stabilizer by offering lower funding in dollars, and rates at least as low as RMB funding. However, it also shows that China took on the role of stabilizer in the second half of the crisis. The post-crisis trend has been for the US to maintain lower rates than China and to continue to serve as the lender of last resort in this respect.
(5.C) Lender of last resort functions: Interbank market.
Between December 2007 and October 2008, the US Federal Reserve created 14 temporary dollar liquidity swap lines with foreign central banks. In December 2007 the Federal Reserve established dollar liquidity swap lines with the European Central Bank (ECB) and the Swiss National Bank (SNB) to enable them to provide dollar funding to financial institutions operating within their jurisdictions. In May 2008 the People’s Bank of China, the Bank of Japan, Bank of Korea and ten ASEAN countries agreed to create a US$80 billion multilateral swap facility known as the Chiang Mai Initiative Multilateralization (CMIM), though this were never used (Sim and Piboontanasawat, 2008). By September 2008 new lines were opened with the Bank of Japan, the Bank of England and the Bank of Canada, the Reserve Bank of Australia, the Swedish Riksbank, the Norwegian Norges Bank, and the Danish Danmarks Nationalbank. By October 2008 the Federal Reserve had provided temporary swap facilities to the Central Banks of Brazil, Mexico and South Korea and the Monetary Authority of Singapore in order to boost dollar liquidity (Torres and Sim, 2008). The temporary dollar swap lines between the Federal Reserve and the Bank of Canada, the Bank of England, the ECB, and the Swiss National Bank, as well as the system of bilateral lines, were subsequently extended until 01 February 2014 (BGFRS, 2012).
In April 2009, the Federal Reserve created foreign-currency liquidity swap lines with the Bank of England, the European Central Bank, the Bank of Japan and the Swiss National Bank. These lines allowed the Federal Reserve to obtain foreign currency injections from foreign central banks that it could then lend to American companies and financial institutions. These temporary foreign-currency liquidity swap lines between the Federal Reserve and the Bank of Canada, the Bank of England, the Bank of Japan, the European Central Bank and the Swiss National Bank were re-introduced in November 2011 and included bilateral swap lines between the foreign central banks.
In sum, in terms of lender of last resort functions, the US provides liquidity to a much larger extent than China because dollars are the world’s most internationalized currency and the expansion of America’s monetary base has a worldwide impact. China, on the other hand, holds the world’s largest stock of dollar reserves and is therefore capable of providing liquidity by supplying other actors with accumulated dollar reserves.
In terms of interest rate liquidity, the Federal Reserve acts as lender of last resort to a much greater extent than the People’s Bank of China. Dollar funding has been lower than RMB funding in the interbank market – both over the long term and during the Great Recession. The US is likely to remain the main lender of last resort for the foreseeable future. A clear follow-the-leader dynamic also emerges with respect to crisis liquidity provision with the US the first to boost liquidity through swap lines.
A critical perspective on Kindleberger’s functions
Having examined Kindleberger’s five functions we note several limitations in their application to the US and China. First, when the US and China provide a market for distressed goods, there is a difference between routine periods and those of crisis. While China has become a major importer and is displacing the US, it mostly imports and processes component parts, assembling them into final goods. The low value added of Chinese imports makes only a limited contribution to soaking up global demand as required by a market for distressed goods, i.e., where traded goods can go in the absence of other markets.
Second, a distinction between routine and crisis periods is also useful in terms of the provision of a counter-cyclical flow of capital. China’s capital exports did not effectively contribute towards stabilization in normal times because they provided the US with huge capital inflows, contributing to the onset of the Great Recession as well as resulting in large trade surpluses and destabilizing global imbalances. However, sizeable capital exports are, conversely, stabilizing in crisis periods.
Third, as mentioned, Kindleberger was a fan of fixed exchange rates, but economists today see benefits with flexible exchange rates for balance of payments adjustments. During a period of crisis, we should expect a flight to safety and therefore greater volatility in the exchange rate of the leading economy. Here too, then, a distinction between crises and normal periods is warranted.
Assessment of findings
An evaluation of the respective behavior of the US and China in regards to these five functions is summarized in Tables 1, 2 and 3. Here we divide our assessment into three categories.
Summary evaluation of functions: leadership, in normal times.
Summary evaluation of functions: stabilizer.
Summary evaluation of functions: trend.
Table 1, on “leadership”, characterizes our assessment of American and Chinese behavior according to each indicator for the balance of the 1996 to 2013 period in normal times as specified earlier. Table 2 addresses the behavior of the two countries as a stabilizer during the three periods of crises. Table 3 reflects the long-term trends: it therefore captures our assessment of the relative trajectory of each country based on their progression over the time period from 1996 up to the present.
As Table 1 demonstrates, the US is the predominant provider of the five functions in routine periods, if not the sole leader. It provides a counter-cyclical flow of capital, coordinates macroeconomic policy and acts as lender of last resort in routine periods when there is no crisis. The US does however share two tasks with China: of providing a market for distressed goods and a stable exchange rate.
A different picture emerges, however, when examining the role of the US as a stabilizer during a period of crisis – the contingency that Kindleberger explicitly addressed in his work. From Table 2 it is apparent that the five functions were far more evenly shared with China. During the Great Recession, which is the most important crisis because of its scope and scale, both the US and China were responsible for providing a market for distressed goods and a counter-cyclical flow of capital, for stabilizing exchange rates and for acting as lender of last resort (even though the US played a predominant role in providing this last function). Of the five functions identified by Kindleberger, the US singlehandedly provided one of these: China, however, played an ever-increasing role, independently providing one function in the period post-dating the Great Recession while sharing the others.
A similar pattern emerges when expanding the analysis to all three crises. The US and China shared the provision of a market for distressed goods and stable exchange rates relatively equitably. The US took on primary responsibility for providing a counter-cyclical flow of capital and lender of last resort. China however, played a contributory role in both dimensions. China’s involvement as lender of last resort is key because it is the specific component on which many IPE scholars focus in discussing America’s indispensable economic role (Helleiner, 2014). The US played an exclusive role only in the coordination of macroeconomic policy.
Finally, the trend lines displayed in Table 3 point unambiguously to even greater sharing of these five functions over the long term (possibly in routine and crisis periods) and even a reversal in the order of responsibility for them. During the Great Recession, the US was solely responsible for stabilization according to one to two functions, the same number as China, while responsibility for stabilization was otherwise shared. Extrapolating the trends, China is poised to take the lead in providing a market for distressed goods, a counter-cyclical flow of capital and stable exchange rates. Based on these findings however we would expect the US to continue to dominate macroeconomic policy coordination and to act as the principal lender of last resort.
Conclusion and implications
At the outset we listed the concerns of this article: to examine the contemporary utility of Kindleberger’s work, the respective roles of China and the US as potential stabilizers in a succession of recent crises, and thus to evaluate whether stabilization functions can be shared or if the world does require one stabilizer in an economic crisis.
Our results suggest that performing stabilization functions, as Kindleberger defined them, may remain key. However, in effect, our empirical findings present a far more mixed picture about either American or Chinese behavior than conventional IR scholars would anticipate – with significant implications for both prevalent theoretical and policy assumptions. Bearing this in mind, we offer three concluding comments.
First, Kindleberger, as quoted at the outset, argued that the global economic system needed one stabilizer during a period of crisis. This assertion has formed one of the central justifications for sustained American global leadership. The seminal work of Keohane (1984) and Snidal (1985, 1991) pointed to the limitations of this proposition. Keohane’s work stressed the potential multilateral role played by regimes as stabilizers “after hegemony”: Snidal’s work suggested that, at least in principle, stabilization functions could be shared amongst two or more actors. Complementing their theoretical critique, we offer a critical empirical test of this “one-stabilizer” proposition in the context of the Great Recession. Our evaluation reveals that Kindleberger’s five functions can and were effectively divided. However, as a significant addendum to both Keohane and Snidal’s work, and to Realist formulations, we found that China and the United States shared these functions in the absence of at least explicit or direct evidence of collusion, cooperation or coercion. Although the overall effectiveness of their performance (both individually and collectively) is open to further assessment, the fact remains that the system remained intact, functioning and was, for at least some parts of the globe, prosperous both during and in the aftermath of the crisis. Our findings therefore challenge a primary assumption of this broad research program, promoting the counterclaim that at least two actors can perform these stabilization functions without direct official cooperation, although signaling and unofficial discussions may lead to degrees of implicit coordination. The IR and IPE literature surveyed in this article depends often, to varying degrees, on the “single stabilizer assumption” to justify claims about American global economic leadership. We do not claim that our findings challenge the entire body of literature in IR and IPE adopting this assumption; but we do claim that it casts doubt on analysis and prescriptions that are contingent on this assumption. We propose in future research to examine the interactions of a stabilizing duopoly and how it influences the dynamics of the global economy. In some respects, even a G–2 framework may be too restrictive in fully capturing leadership dynamics in the modern international economy. Evaluating the role of the European Union and emerging powers is a promising avenue of research, although circumstances dictated they played no major role as stabilizers during the Great Recession.
Second, the other hypothesis we tested is that China shared the role of crisis stabilizer with the US during the Great Recession. Our findings support this hypothesis. China played an increasingly important role across the three crises culminating in the Great Recession. These efforts to sustain the system as a stabilizer are inconsistent with two contrasting characterizations. The first is the depiction of China as either a predatory challenger – a foundational proposition of power transition theory (see for example, Efird et al., 2003; Goldstein, 2007; Mearsheimer, 2001; Organski and Kugler, 1980: 19–20, 23; Rapkin and Thompson, 2003; Schweller and Pu, 2011: 105; Tammen and Kugler, 2006; Tammen et al., 2000) The second is the Liberal depiction of China as a possible free rider (Fravel, 2010; Johnston, 2003).
Indeed, the evidence we have presented supports a markedly different story to classic expectations: that both the US and China performed invaluable stabilization functions during the greatest economic crisis in a century. Our findings thus have potential implications for two contrasting approaches to the study of US–China relations.
Offensive Realists examining our findings may argue that it provides evidence in support of their position by confirming their greatest fear: that the trend is away from long-term American leadership and towards China. But their fears can only be regarded as viable if they understate the fact that the Chinese did not seize the opportunity to challenge American leadership and change the “rules of the game” during the Great Recession as Offensive Realist theory might predict. China preferred to support the status quo. As Alistair Ian Johnston suggests, “For a concept at the core of international relations theorizing, it is disturbing how little thought, with the exception perhaps of Gilpin, has gone into determining whether a state is status quo or revisionist across the totality of its foreign policy preferences and actions” (Johnston, 2003: 10). Robert Gilpin broke down the rules of the game into operable components: the distribution of power, the hierarchy of prestige (which, however, tends to be co-terminus with the distribution of power for Realists), and “rights and rules that govern or at least influence the interactions among states.” (Gilpin, 1981: 34). China’s leadership supported the system in crisis and, realistically, made no major demands in terms of “rights and rules”. Evidence to the contrary is cursory: even the marginal shift in voting rights in the IMF that followed the Great Recession, for example, did not especially favor China (Woods, 2010).
However, our findings pose a comparable challenge to some Liberal scholars’ characterization of China. While many Liberals anticipate the prospect of an increasingly integrated China in the future, acting within the confines of the current system, they also often anticipate sustained American global economic leadership and accuse China of being an “irresponsible stakeholder” – a free rider – in contrast to benevolent and generous American leadership (Brooks and Wohlforth, 2008; Foot and Walter, 2011; Ikenberry, 2008; Kang, 2007; Kirshner, 2007; Moore and Dixia, 2001; Zhu, 2008;). The idea that American benevolence is what distinguishes the United States from other nation states is an assumption we have criticized, more generally, before (Norrlof, 2008, 2010; Reich and Lebow, 2014). There is little in our findings to support the benevolence proposition in the context of global financial crisis management. The US did not act more responsibly than China during the critical period of the crisis. In contrast, Chinese behavior, while self-interested, cannot be characterized as “free-riding” according to these data. In its broadest and most vernacular sense, China proved just as “responsible” during the crisis as did its American counterpart.
Our third and final point is that our findings challenge recent work that focuses on the effective functioning of global institutions in explaining the ability of the global economy to traverse the Great Recession. Daniel Drezner has argued, for example, that, “the system worked” primarily because of the ways in which global institutions functioned. Rather, we focus on a state-based explanation for the stabilization of the global economy, albeit one that emphasizes the behavior of two nation states acting without explicit cooperation – one of which, namely China, demonstrated little or no regard for the policies of global institutions (Woods, 2010). 12
Our article suggests interesting future avenues for research. Some are theoretical, such as those regarding the dynamics of a two-actor stabilization system in which each actor wants to preserve the current system but acts autonomously in a self interested manner without explicit mechanisms of cooperation – a significant contrast to earlier formulations based on cooperation or coercion. Others are more technical, such as the relative weight that should be given to each of Kindleberger’s functions or the ways that the functions work in tandem or – as at least two generations of scholars have contended but not necessarily evaluated – if all of these functions remain relevant.
Our examination of Kindleberger’s criteria supports the idea of the US as the single most significant actor in the international system during normal times, a position shared with China during a period of crisis, and predicts a more complex future. If Kindleberger’s criteria are indeed correct, the assumption of a single country, the United States, acting as a stabilizer is empirically mistaken and with it is an ensuing welter of research based on the assumption of the virtues of American primacy. Consequently, we would argue that international relations theory is in need of drastic revision to reflect how hierarchy relates to responsibility in times of crisis (Lake, 2009), and how leadership shifts actually do occur – without challenges, coercion or the dénouement of the current global capitalist system. If Kindleberger’s functions are irrelevant to governance and responsibility, major theories that have assumed the criteria to be relevant for explaining systemic leadership and crisis stabilization are in need of a critical reassessment.
Footnotes
Acknowledgements
We are grateful to Jerry Cohen, Scott Gates, Joseph S Nye, Nita Rudra and Jun Xian for helpful comments on an earlier version of this paper and thank Meghan Sullivan for excellent research assistance.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.