In the discussion of our contemporary economic disease, the Great Depression analogy refuses to go away. Almost every policy-maker referred to conditions that had ‘not been seen since the Great Depression’, even before the failure of Lehman. Some even went further – the Deputy Governor of the Bank of England notably called the crisis the worst ‘financial crisis in human history’. In its April 2009 World Economic Outlook, the IMF looked explicitly at the analogy not only in the collapse of financial confidence, but also in the rapid decline of trade and industrial activity across the world.Footnote ² In general, history rather than economic theory seems to offer a guide in interpreting wildly surprising and inherently unpredictable events. Some observers, notably Paul Krugman, have concluded that a Dark Age of macroeconomics has set in.Footnote ³

This article examines three areas in which analogies have been made between the interwar depression and the financial crisis of 2007, which reached a dramatic climax in September 2008 with the collapse of Lehman Brothers and the rescue of AIG: they can be labeled macroeconomic, microeconomic and geopolitical. First, the article considers the story of monetary policy failures. Second, there follows an examination of the microeconomic issues concerned with bank regulation and the reorganization of banking following the failure of one or more major financial institutions and the threat of systemic collapse. Third, the article turns to the issue of global imbalances and asks whether there are parallels that might be found in this domain too between the 1930s and the events of today.

I

Almost every contemporary use of the Depression analogy takes the year 1929 as a reference point. But there are really two completely different pathologies during the Great Depression, which involve different diagnoses and different cures.

The first, and the most famous, pathology is the US stock market crash of October 1929. No other country had a stock market panic of the magnitude of the American one, in large part because no other country had experienced the euphoric run-up of stock prices that sucked large numbers of Americans, from very different backgrounds, into financial speculation. The second sickness, contagious banking panics, was decisive in turning a bad recession into the Great Depression. A series of bank panics beginning in October 1930 in the United States converted a not unusual recession from 1929 to 1930 into a serious slump. Through the fixed exchange rate gold standard the US depression also affected the rest of the world. Events took a turn for the worst after the collapse and rescue of the Creditanstalt bank in Vienna in May 1931 and a major banking crisis in Germany in June. This spread financial contagion to Great Britain, to France and back to the US.

The 1929 panic has dominated a great deal of the analysis of the Depression for two rather peculiar reasons. First, no one has ever satisfactorily been able to explain the collapse of the market in October 1929 in terms of a rational explanation, in which market participants reacted to a specific news event. So the crash presents an intriguing intellectual puzzle, and economists can build reputations on trying to find innovative accounts. Some people just conclude that markets are simply irrational. Others have argued that investors might have been able to foresee the Depression, or that they were pondering the likelihood of protectionist reactions in other countries to the American (Smoot Hawley) Tariff Act, which had not yet even been cast in its final form.

The second reason that 1929 has been popular with academic and political commentators is that the aftermath of the collapse provides a clear motive for taking particular policy measures. Stock exchange collapses or the end of asset bubbles do not necessarily lead to prolonged recessions of deep depression. In October 1987 and again in March 2000 sharp stock market collapses triggered both an extension of liquidity by the central bank and fiscal easing. Keynesians thought that government fiscal demand can stabilize the expectations of the market, and thus provide an overall framework of stability. Monetarists saw monetary stability as the key to avoiding dramatic output contractions. Much of this debate has focused on the United States: in other countries, especially debtor countries, the gold standard constrained monetary policy so that it is hard to speak of policy options. The only country where there was equivalent room for maneuver to the United States is France.

The Great Contraction of 1929-33 in the United States, during which prices, real output and money supply declined by about a third, and which spread to the rest of the world, was precipitated by policy failures at the Federal Reserve. A tight monetary policy to kill stock market speculation in 1928 led to a recession beginning in August 1929. This policy was based on the real bills view that stock market speculation would lead to inflation, a bust and then deflation. The stock market crash in October exacerbated the downturn but did not cause the depression. The failure of the Fed to follow its mandate from the Federal Reserve Act of 1913 to act as lender of last resort and to allay a series of four banking panics beginning in October 1930 led to the serious downturn that followed. The Fed adhered to the flawed Burgess Riefler doctrine,Footnote ⁴ which viewed low levels of its borrowed reserves (i.e. discount window borrowing) and short-term interest rate indicators as signs of monetary ease and hence did not act. In addition, some Fed officials believed in the liquidationist doctrine and saw bank failures as beneficial. A major hike in the discount rate in the fall of 1931 to protect the dollar after sterling exited from the gold standard added fuel to the fire.

Recovery began in March 1933 with Roosevelt's banking holiday, ending the fourth banking panic. The nation's banks were closed for a week during which an army of bank examiners separated the insolvent from the rest. Insolvent banks were closed, ending the uncertainty driving the panic. This action was quickly followed by FDR taking the US off the gold standard (abrogating the gold clauses and prohibiting private gold ownership) in April, Treasury gold (and silver) purchases designed to raise gold prices and prices in general, and formal devaluation of the dollar by close to 60 percent in January 1934. These policies produced a big reflationary impulse from gold inflows which were unsterilized, passing directly into the money supply. They also helped convert deflationary expectations into inflationary ones.Footnote ⁵ Also of key importance in preventing future banking panics was the institution of federal deposit insurance (FDIC) in the Banking Act of 1933, which went into effect on 1 January 1934.

The recovery of 1933–41 in the United States was largely driven by gold inflows (initially reflecting Treasury policy and the devaluation, later reflecting capital flight from Europe as war loomed). Expansionary fiscal policy, despite the conventional wisdom, played only a minor role in the recovery of the 1930s.Footnote ⁶ Recovery was impeded somewhat by New Deal cartelization policies like the NIRA, which in an attempt to raise wages and prices artificially reduced labor supply and aggregate supply.Footnote ⁷ Over the period 1933-7 output increased by 33 percent.

The Federal Reserve was largely passive in the 1930s. Along with the bankers, it had been blamed by the Roosevelt administration for the failures of the 1920s and early 1930s. Major reforms in the Banking Acts of 1933 and 1935 greatly increased the powers of the Federal Reserve Board in Washington at the expense of the Reserve banks and especially the New York Fed. Despite its increase in power, the reconstituted Board of Governors under Chairman Mariner Eccles was passive and largely subservient to the dictates of Treasury Secretary Morgenthau. The Fed in the 1930s continued to follow the same precepts as it did in the 1920s and early 1930s. Its policy indicator continued to be the level of free reserves (excess reserves less borrowings from the Fed). In the 1930s borrowed reserves were negligible, so excess reserves became the indicator. As the decade wore on, member banks largely absorbed the gold inflows into excess reserves, held as a precaution against a repeat of the type of turbulence experienced in the early 1930s. By 1935 excess reserves amounted to 50 percent of total reserves. Fed officials increasingly viewed the build-up of excess reserves as a threat to future speculation and inflation. They also saw the presence of sizable excess reserves as preventing them from future tightening. Similar concerns have been voiced about the build-up in bank excess reserves in 2008-9. According to the Burgess Riefler doctrine which prevailed at the Fed, the way the Fed could control interest rates was by forcing banks to borrow from the Fed. Once borrowed reserves were less than the open market portfolio, then open market sales could force the banks to borrow. Banks would then want to reduce their indebtedness by contracting their lending.Footnote ⁸

The recession of 1937–8. The recovery was interrupted by a serious recession (the third worst of the twentieth century) from May 1937 to June 1938. Friedman and Schwartz and, more recently, Meltzer and others attribute the recession to a serious policy mistake by the Federal Reserve.Footnote ⁹ Mounting concern by the Fed over the build-up in excess reserves in member banks led the Board to double reserve requirements in three steps between August 1936 and May 1937. The rationale for this action was to restore the Fed's control over monetary policy and remove the inflationary threat posed by the excess reserves. The Fed used the blunt instrument of raising reserve requirements rather than conducting an open market sale of securities because excess reserves exceeded the Fed's portfolio of securities and sales would reduce the income earned from it. According to Friedman and Schwartz, the banks were holding excess reserves as a precaution against a repeat of the banking panics of the 1930s. When the Fed locked up these reserves the banks cut back on lending and sold earning assets to restore the precautionary cushion they had held. The Fed's contractionary policy action was complemented by the Treasury's decision in late 1936 to sterilize gold inflows in order to reduce excess reserves. These policy actions led to a spike in short-term interest rates and a severe decline in money supply precipitating a 5 percent decline in real GDP.

Other explanations given for the recession of 1937-8 include: a tightening of fiscal policy when the administration ended a generous veteran's bonus, hiked income tax rates and imposed a tax on undistributed profits; gold hoarding brought about by fears of another dollar devaluation coupled with a boost to money wages by the Wagner ActFootnote ¹⁰ and a switch back from inflationary to deflationary expectations.Footnote ¹¹

The recession ended after FDR in April 1938 pressured the Fed to roll back reserve requirements, the Treasury stopped sterilizing gold inflows and desterilized all the remaining gold sterilized since December 1936, and the administration began pursuing expansionary fiscal policy. The recovery from 1938 to 1942 was spectacular, output grew by 49 percent fueled by gold inflows from Europe and a major defense build-up.

The liquidity trap. The 1930s were characterized by very low interest rates. Short-term rates were close to zero through much of the decade. Long-term rates were close to 2 percent. The traditional Keynesian view has been that monetary policy was impotent because the US economy was in a liquidity trap. Like the 1930s, a Federal Funds rate in 2008 close to zero (the zero lower bound) has again raised the issue of policy impotence.

Subsequent research by Brunner and MeltzerFootnote ¹² found no evidence for the liquidity trap. There was a spectrum of rates well above zero throughout the 1930s and the Fed could just as easily have bought securities other than short-term Treasury bills.Footnote ¹³ The real problem was not that Fed policy didn't work but rather that the Fed was unwilling to use the tools that it had to conduct expansionary monetary policy because it feared a resurgence of asset market speculation and inflation.Footnote ¹⁴

Lessons for today. The history of the 1930s experience has several lessons for the present discussion over the policies that the Fed could follow to ensure a rapid recovery without engendering inflation.

The first lesson is that the Fed, like its predecessor seventy years ago, has the tools to reflate the economy and to prevent a resurgence of inflation. In the 1930s the Fed was only a minor player in the recovery because it was reluctant to use expansionary open market purchases for fear of rekindling speculation and inflation. It was not in reality stuck in a liquidity trap or hampered by the zero lower bound because the rates on many securities were positive and the Fed could have purchased them. Instead, the Treasury through its policies towards gold and the consequence of devaluing the dollar did more of the heavy lifting to promote recovery.

In the recent crisis the Fed's policy of sterilizing the effects on the monetary base of its diverse liquidity operations through much of 2008 (until September) made monetary policy tighter than it had to be and likely exacerbated the recession which began in December 2007.Footnote ¹⁵ However, since October 2008 the base has greatly expanded and the policy adopted in January 2009 of quantitative easing by purchasing long-term Treasuries and mortgage-backed securities can be viewed as a replay of the expansionary Treasury gold policy of the 1930s.

Second, the Fed will eventually have to tighten as the economy recovers and excess capacity is reduced. Some have raised the fear that this could produce a repeat of the recession of 1937-8 were the Fed to attempt to reduce the excess reserves and the banks (still gun-shy from the recent crisis) to scramble to replace them. This should not be a problem for a number of reasons. First, the excess reserves were built up in the two eras under very different Fed operating procedures. In the 1930s, the Fed could not target the interest rate as it had done in the 1920s because the banks were reluctant to borrow reflecting a stigma from doing so. Moreover, the build-up of excess reserves was a consequence of the gold inflows and, given the Fed's preferred operating procedures, created a problem for it.

Today the Fed follows an interest rate target and it can pay interest on reserves (IOR). The build-up of reserves reflected sterilization of the Fed's liquidity operations using interest on reserves (when the federal funds rate was close to zero), as the mechanism to get banks to hold them. Were the Fed to wish to tighten, it could separate its monetary policy operations from its liquidity policy by changing the spread between the funds rate and the IOR.Footnote ¹⁶ Unlike the Fed of the 1930s, today's Fed can use reverse repos or open market sales of its long-term securities to do the tightening. Were it to wish to reduce excess reserves to encourage banks to lend, it could pay negative interest on reserves, as was done recently by the Riksbank in Sweden.

The main concern for today is not that the Fed can not exit from its present strategy, because it can; but that when it exits and begins tightening, if unemployment were still to be high and were to begin to rise again in the face of the tightening, the Fed would come under political pressure to abandon its efforts and cave in under the pressure. In that case, inflationary pressures would build up as the markets and the public began to doubt the Fed's resolve. This is what happened in 1966 and 1969 under William McChesney Martin and in 1973 under Arthur Burns, leading to the Great Inflation. Moreover, if the recovery turns out to be rapid, as was the case in the 1930s (and virtually all the deep recessions in the twentieth century),Footnote ¹⁷ then inflationary pressure may build up sooner than many have expected . In such a scenario, does the current Fed have the resolve to follow through on an anti-inflationary policy?

II

Banking collapses played a crucial role in the deepening of the global crisis in 1931. Unlike the United States, where banking was highly localized, continental European economies were dominated by financial systems in which a small number of very large banks dominated the economy. In Austria, where the crisis began in May 1931, the Creditanstalt controlled some 60 percent of Austrian firms through ownership stakes.Footnote ¹⁸ The failure or potential failure of very large financial institutions thus posed a major policy problem.

The collapses were the result of the shocks of the international depression imposed upon bank weakness in countries that had been wrecked by the aftermath of bad policies that produced inflation, hyper-inflation and a destruction of banks' balance sheets. An intrinsic vulnerability made for a heightened exposure to political shocks, and disputes about a central European customs union and about the postwar reparations issue was enough to topple a house of cards.

But finding a way out of the damage was and is very tough. Unlike 1929, there are no obvious macroeconomic answers to financial distress, particularly when it involves institutions that are deemed to be ‘too big to fail’. Some famous macroeconomists, including Larry Summers, the current chief economic thinker of the Obama administration, in consequence tried to play down the role of financial sector instability in causing depressions. Robert Lucas's claim in 2003 that the ‘central problem of depression-prevention has been solved’ is one of the central pieces of evidence for Krugman's onslaught. The answers to financial stress lie in the slow and painful cleaning-up of balance sheets; and in microeconomic restructuring, which cannot be solely imposed from above by an all-wise planner but also requires many businesses and individuals to change their outlook and behavior. The improvement of regulation and supervision, while a good idea, is better suited to avoiding future crises than dealing with the consequences of a catastrophe that has already occurred.

Banks in 1931 were vulnerable as a result of poor monetary policy, and they were victims of monetary deflation.Footnote ¹⁹ But there were plenty of specific issues which long antedated the collapses of the early 1930s.Footnote ²⁰ They are the result of specific design features of the financial system that could not simply be corrected by macroeconomic policy, whether monetary or fiscal. US banking was highly localized, and thus vulnerable to geographically limited shocks (such as the agricultural depression); while larger nationwide banking in Canada was much more resilient. Banks in many debtor countries in South America and Central Europe accumulated mismatches between assets (in local currency) and liabilities (in dollars or other key currencies), that made for a vulnerability to currency turmoil. Universal banks suffered large losses on their shareholdings, and as their capitalization fell, cut back on their lending. Some British banks (the so-called merchant banks) had heavy overseas exposures that made them vulnerable to foreign crises.Footnote ²¹

The consequence of the long academic and popular discussion of 1929 is that people have come to the expectation that there must be easy answers. But the collapse of Lehman Brothers in September 2008 was a 1931-like event, the failure of a large financial institution. The answers required are less obvious than in the domain of monetary or fiscal policy, where lessons of the Great Depression are much clearer.

One of the striking features of the Depression analogy is how many of the answers regarding the banking sector are popular again today: in particular, the provision of state guarantees to attempt to revive the interbank market and bank lending; recapitalization of banks with public money; and the establishment of ‘bad banks’ to take problematic assets off banks' balance sheets. All of these policy responses were tried in the 1930s, most notably in the epicenter of the Central European collapse, in Germany.

Some of the initiatives that the German government took had a quite modern ring to them. Indeed, this was an area in which the German government appeared to act swiftly in order to implement a crisis management strategy. First, the government reorganized the banks, merging the two weakest ones, Danat and Dresdener Bank, that had been at the origin of the banking collapse, and injecting government money into all of them. Initially, the government had tried hard to get private money as well, and there were intense negotiations with the leading figures of the powerful Rhine–Ruhr steel lobby. In the end the business leaders only agreed if the government would put in more money, and if the government advanced them the sums that they were supposed to invest in the recapitalization of Danat Bank. By 1932, 91 percent of the Dresdner Bank's capital, 70 percent of Commerzbank's and 35 percent of Deutsche Bank's was in public ownership.

Second, the German central bank (the Reichsbank) pushed for a new institution which would allow it to discount bills from banks which could not be traded because the interbank market had stopped operating. This institution, named the Akzept- und Garantiebank, was established with breathtaking speed. It was given a public guarantee in order to provide the additional signature that made bills eligible for Reichsbank lending (rediscounting).

Third, the Reichsbank eventually (in December 1932) created what would now be called a ‘bad bank’ to take over troubled assets whose prices no longer corresponded to the value at which they were set in the banks' balance sheet. Two new institutions would take assets off firms' and banks' balance sheets: the first, the Deutsche Finanzierungsinstitut AG took over up to three-quarters of the bad assets of a bank, but required an annual amortization at 3 percent. The second, the Tilgungskasse für gewerbliche Kredite, required a much lower rate of servicing, only 1 percent, for an initial three-year period, followed by higher rates as economic recovery set in.

Bailouts are inherently controversial, because they distribute public money in an arbitrary way, to one recipient rather than another. In the United States, Herbert Hoover's innovative Reconstruction Finance Corporation of 1932 quickly ran into problems because of this issue: it turned out that the credits were going to banks, farms and businesses that were well connected with Republican politics. Germany offers an even more dramatic example of this kind of problem. As part of the bank bailout in the aftermath of the 1931 crisis, 2.5 m Reichsmarks was put into a small Berlin institution, Hardy & Co., which was a subsidiary of the Dresdner Bank. This money was primarily intended to flow into the electoral campaign coffers of Paul von Hindenburg, the veteran World War I commander who had been elected President of Germany and was standing for re-election in 1932.Footnote ²²

In the fragile situation of Weimar Germany, the bailout that was at the center of the government's response to the banking crisis ran into every kind of objection. The claim that the government had been engaged in the ‘socialization of losses’ became an important part of the turbulent electoral campaigns of 1932. In order to get support from the Akzeptbank, banks had to demonstrate that ‘important economic interests’ were at stake, and in practice the majority of Akzeptbank credit went to the savings banks (Sparkassen). It was also used to support enterprises in strategically vital areas, notably Silesia. The special issues involved in the support of Silesian industry, and the fear of an opportunistic takeover by foreign issues, led to the Chancellor Heinrich Brüning's most problematical and indeed scandalous rescue operation, the so-called Gelsenberg purchase concluded on the last day that Brüning and his Finance Minister Hermann Dietrich, the driving force of this bailout, were in office. In this transaction, the government, which as a result of the banking crisis had become Flick's largest creditor, bought out Flick's interest in the steel giant Vereinigte Stahlwerke. Dietrich's former State Secretary Hans Schäffer referred to the operation as ‘extreme stupidity’.

We can see the same crippling effects of a bailout in the comparatively much more expensive and extensive case of Austria, where the collapse of the Creditanstalt in May 1931 was the precipitant of the more general Central European financial collapse. The government's answer involved taking over the bank, and eventually merging it with other weakened Austrian banks, the Wiener Bankverein and the Niederösterreichische Escompte Gesellschaft. The subsidy was expensive, amounting to 9 to 10 percent of GNP, substantially less than the cost of bailouts for Mexico or Japan in the 1990s, but comparable to the projected costs of the 2008 bailout in the UK and Ireland. Since the Creditanstalt held major stakes in some 142 Austrian firms, it meant that the government through the bank was in effect running most of Austrian business. Similarly to modern bailouts in emerging markets, it was also accompanied by massive corruption, the revelation of which became the stock-in-trade of the opposition Nazi movement in Austria. Then, as now, there was massive public hostility to the idea of a bailout, in that it appeared to be a form of support for the institutions and people who really bore the responsibility for the crisis.

The cost of bailouts, even when they seemed to have been administered promptly and with high efficiency as in the German case, thus exceeded the simple fiscal arithmetic. They brought the state into a series of contentious microlevel decisions on the health of particular enterprises and on the fate of individual bank directors. Given the poisonous ideological backdrop of anti-Semitism in the context of Central Europe in the 1930s, it is unsurprising that this radical doctrine was fanned by the character of the government's response to banking crises, and that both in Germany and more explicitly in Austria a process of expropriating Jewish property (‘aryanization’), which was at first called Germanization or Austrianization, set in even before the Nazis took power in those countries. The episodes of managing bank failures in retrospect look like the beginning of a process of state-domination, corruption and even racial persecution that would roll on like an ever more menacing snowball.

The politics of bank and industrial bailouts after 2008 have already raised fears of a new financial and economic nationalism, as governments become more directly involved in the micromanagement of the economy. Banks in state ownership with a substantial degree of public investment – Citigroup, Lloyds-HBOS, RBS, Commerzbank – have cut back on foreign activities and sold foreign assets, at least in part because of government pressure that taxpayer money should not be used for the benefit of foreign borrowers. Economic nationalism is even more evident in the debate about government rescues of the automobile industry, where domestic jobs are protected at the cost of foreign jobs in an industry dealing with global overcapacity.

One key problem at the heart of both the 1931 crisis in Europe and 2008 in the US and Europe was the doctrine of ‘too big to fail’. It was born in the aftermath of the Latin American debt crisis of 1982, which threatened the solvency of almost all financial institutions in the industrial countries. In 1984, the doctrine was applied to justify the decision to bail out Continental Illinois, the fourth biggest US bank, which was insolvent. As banks grew in the 1990s and 2000s, and their interconnectedness increased, the doctrine evolved and was augmented by an argument about banks being ‘too interconnected to fail’. In 2008, the doctrine contributed to the worsening of financial crisis, as the belief that large commercial banks would not be allowed to fail was extended to investment banks with the rescue of Bear Stearns in March 2008. Then in September, when Lehman Brothers was allowed to fail and AIG was rescued, the resulting confusion led to panic. Too big to fail has also hampered the recovery by preventing the use of the good bank/bad bank solution (that had been used so successfully in the past by Sweden and other countries) towards Citigroup, Bank of America in the US and some big banks in Europe: RBOS, Lloyds-HBOS, UBS. In consequence, governments have taken large shares in financial institutions in order to recapitalize them, a move analogous to what happened in 1931 in Germany.

III

There is one further way that the aftermath of Lehman looks highly reminiscent of the world of depression economics. Austrian and German bank collapses would not have knocked the whole world from recession into depression if those countries had simply been isolated or self-contained economies. But they had built their economies on borrowed money in the second half of the 1920s, with the chief sources of the funds lying in America. But after 1931, they could not recapitalize themselves via capital inflows. The analogy of that dependence is the way in which money from emerging economies, mostly in Asia, flowed into the US in the 2000s, and an apparent economic miracle was based on the Chinese willingness to lend. The bank collapses in 1931, and in September 2008, shook the confidence of the international creditor: then the United States, now China.

In the 1930s, the United States largely stopped lending to Europe. China has not undertaken a similar reversal, and China's reserves are still being used to finance the massive US budget deficits. But the crisis did prompt a reassessment of Chinese strategy, with a shift of emphasis to domestic investment as well as to the use of foreign exchange to acquire strategic assets (commodity and energy suppliers as well as farmland) outside the industrialized world, largely in Africa and in South America.

After the crisis, a question arose as to whether the flows would resume. In the 1930s, they did not and the flows of the 1920s were reversed, with movements out of Europe and South America and into the United States after 1933. The abrupt reorientation of capital movements led KindlebergerFootnote ²³ to argue that the Great Depression could have been much milder if Americans had been willing to continue to lend to Europe or, in other words, if the US had taken on the role of international lender of last resort.

The diversion of capital movements in the 1930s was also a response to trade protection, in that the former borrowers could no longer hope to service their debt through exports. In the 2000s, the absence as yet of a significant amount of trade protection may thus protect the continuation of capital market openness.

Nevertheless, there are many reasons to think that, as in the 1930s, finance will become more national in the aftermath of the crisis. The logic of bank rescues brings an immediate and increased pressure on financial institutions to concentrate their activities in a national setting, as tax payers and politicians are reluctant to subsidize institutions that lend to far-away borrowers (whose products may be in competition with those of domestic manufacturers). The large European bank rescues have been accompanied by substantial pressure to contract exposure to non-domestic borrowers, and to sell-off foreign affiliates in order to increase capitalization.

The retreat from financial globalization also marked the intensification of the crisis. The first stage of the recent financial crisis, between the spring of 2007 and September 2008, had been much gentler because many troubled financial institutions recapitalized themselves by selling stakes to Asian or Gulf SWFs. A crucial development was the collapse of negotiations for such a sale between Lehman and China Investment Corporation and Korean investors; and after September 2008, emerging market SWFs no longer wished to acquire stakes in western financial institutions.

China and other emerging markets are likely to continue to play a major role as exporters of capital, because of their high savings rates. There is no doubt that China has become a major presence in international financial markets. By the beginning of 2009, the largest three banks in the world by market value were Chinese: the Industrial and Commercial Bank of China, the China Construction Bank and the Bank of China. In addition, because countries with high reserves avoided the crisis of 2008 and therefore have justified their continued build-up of reserves, which continues the imbalances, there is still an open question as to how the surpluses will be employed. According to one scenario, disillusion with past investments in advanced economies and worry about the extent of the exposure of short-term Treasury paper will lead China and other surplus countries to funnel their resources into the purchase of land, mineral resources and other strategic sources in poor and politically malleable areas of the world. A reluctance of foreign investors to buy US government paper would push up yields and dramatically increase the cost of funding US (and other industrial country) debt.

Today there may be plenty of reasons why the Chinese may be tempted to pull back from their engagement with the old industrial economies and with the United States. In fact, the external political logic carries echoes of 1931, when American banks, investors and the US government did not want to pour in good money after bad to Europe. Some of the arguments that are currently reverberating around Beijing are very reasonable: there is a great deal of uncertainty, and the SWFs might lose a lot of money. CIC would have initially lost some money had it taken a stake in Lehman in the summer of 2008. Some lines of thought are more emotional: might not 2008 be a payback for the American bungling of the 1997-8 East Asia crisis? Many people in many countries will interpret a crisis that unambiguously began in the United States, but affects some other countries more harshly, as evidence of a fundamentally malign US plan.

Crisis tends to heighten national security concerns. The Chinese search for a replacement of the US dollar by a synthetic reserve currency is driven by a political backlash against the perceived iniquities of US financial and economic preeminence.

As in the 1930s for the United States, the search for a more national or autarkic Chinese route may be costly. There are substantial domestic vulnerabilities as well as strengths. There is still an undoubted dynamism in the Chinese economy. The strength of the domestic market meant that growth continued, albeit at a reduced pace, in the grim circumstances of 2008 and 2009. In late 2008, China announced one of the largest national stimulus packages of 4 trillion RMB (or $585 bn). Unlike most of the major industrial countries, public debt is quite limited. Even the gigantic spending program will only push it up to about a quarter of GDP, so there is fiscal room for further infrastructure projects.

At the same time, there are a myriad of domestic reasons why China might be expected to be precarious. They are financial, social, demographic and political. The Chinese banking system is still quite opaque, and may still have to wrestle with the legacy of problems of the 1990s, in particular bad loans to big state-owned corporations that were the consequence of a political logic of directed credit. China is investing large amounts in education, but it may be more difficult to make a creative and innovative society that replicates the dynamism of the United States in the second half of the twentieth century (which was in large part fed by openness, including above all openness to immigration). There is a problem of ageing and even an anticipated demographic decline after the 2040s as a legacy of the one-child policy, and a major imbalance between a surplus of young males and an artificially reduced female population. An authoritarian though reformist regime may find it harder to respond flexibly to popular demands, and may be prone to try to mobilize a reactive nationalism to fend off challenges to its authority.

The pressure to engage in large-scale fiscal stimulation is also likely to alter the balance of China's economic development. Even before the outbreak of the economic crisis, there were two alternative models of Chinese economic development. The first was the rural, family and small business-based boom of the 1980s, and it laid a solid foundation for China's modern economic miracle. But by the 1990s, some of the private-sector growth was being choked off by a rival vision of economic growth, built around prestige projects and the large state-owned enterprise sector. At the same time as Shanghai impressed many commentators as the most modern city in the world, analysts of the Chinese economy suggested that it was one of the least entrepreneurial cities in China.Footnote ²⁴ The new stimulus package with its heavy emphasis on infrastructure investment is likely to push the balance of Chinese development more decisively in this latter direction.

China thus has plenty of reasons why it might want to close itself off to the forces of globalization, as the United States did in some policy areas (especially regarding immigration and trade in the 1920s and in finance during the 1930s). Once again, the experience of the 1930s seems to hold some unattractive precedents. The United States felt uncomfortable with the international institutions of the interwar period, in part because they were aligned with the interests of the old hegemonic power, Britain. The League of Nations looked as if it was in practice a tool of British power. Similarly, in the modern context China worries about whether it is adequately represented in international institutions. Its influence in the IMF and the WTO clearly does not correspond to its real position in the world economy, and to the role that China could play in economic stabilization. Reforming international institutions in order to encompass the geopolitical shift is thus a key issue in deciding whether the geopolitical alterations will be crisis-ridden, abrupt and disruptive, or whether a more gradual and peaceful path of adjustment can be achieved.

IV

There are many lessons from the Great Depression that can and should be learnt in respect to the management of our current crisis. The most important one – where the lesson to be drawn is most obvious – is concerned with the avoidance of the monetary policy error of not intervening in the face of banking crises. The policies of the major central banks – the Federal Reserve, the European Central Bank, the Bank of England – suggest that this is a lesson that has been in the main learnt. However, the Fed, after expanding liquidity in the fall of 2007, then followed too contractionary a policy in the first three quarters of 2008, which may have exacerbated the recession that began in December 2007. Some major economies, notably the United States and China, have also embarked on large fiscal stimulus programs although the jury is still out on their effectiveness.

Other lessons are more problematical. Both the lesson about the slowness and the painfulness of bank reconstruction, and the lesson about dependence on a large external provider of capital, are unpalatable. Limiting the size of banks that are too big or too interconnected to fail is a major political problem, especially as such institutions constitute a powerful lobbying force. The current strategy of guaranteeing banks, but also deposits and a broad range of other liabilities, is likely to encourage a further extension rather than a roll-back of the too-big-to-fail doctrine.

For a long time, it was much easier to repeat the soothing mantra that collectively the world community has learned how to avoid a 1929-type of collapse, and that the world's central banks in 1987 or 2001 clearly showed that they had learned the right lesson. It is undoubtedly meritorious of governments to stabilize expectations, and to prevent a worse downward spiraling of crisis. But policy-makers and their advisers will create inappropriate expectations when some simple policy proposals are built up as the basis for the hope that they alone can guarantee recovery. This may not matter if the rapid recovery in the US in the third quarter of 2009 and the even more rapid growth of China and other east Asian countries makes likely the possibility of a V-shaped recovery. As argued recently by Mussa,Footnote ²⁵ all of the deep recessions in US economic history, including 1929-33, were followed by rapid recoveries driven largely by market forces. But it will matter if another possibility prevails in which there are more echoes of the 1930s, that the woes of the financial system and the inadequacy of bank lending will act as a damper for a long time.