1

Financial Crisis: a Hardy Perennial

The years since the early 1970s are unprecedented in terms of the volatility in the prices of commodities, currencies, real estate, and stocks. There have been four waves of financial crises; a large number of banks in three, four, or more countries collapsed at about the same time. Each wave was followed by a recession, and the economic slowdown that began in 2008 was the most severe and the most global since the Great Depression of the 1930s.

The first crisis wave was in the early 1980s when Mexico, Brazil, Argentina, and ten other developing countries defaulted on their $800 billion of US dollar-denominated loans. The second wave occurred in the early 1990s and engulfed Japan and three of the Nordic countries – Finland, Norway, and Sweden. The Asian Financial Crisis that began in mid-1997 was the third wave; Thailand, Malaysia, and Indonesia were involved initially and subsequently South Korea, Russia, Brazil, and Argentina tumbled. In retrospect the financial crisis that impacted Mexico during its presidential transition at the end of 1994 was the forerunner for the crisis in Southeast Asia thirty months later. The fourth wave began in 2007 and was triggered by declines in the prices of real estate in the United States, Britain, Spain, Ireland, and Iceland – and then by declines in the prices of the bonds of the Greek, Portuguese, and Spanish governments.

Each wave of crises followed a wave of credit bubbles, when the indebtedness of similarly placed groups of borrowers increased at a rate two or three times higher than the interest rate for three, four, or more years. Usually these borrowers used the money to buy real estate – homes and commercial properties. However, the first wave of credit bubbles involved rapid growth in the loans from the major international banks to the governments and to government-owned firms in Mexico and other developing countries that continued for nearly ten years. Japan was the key country in the second wave of bubbles, real estate prices and stock prices increased by a factor of five to six in the 1980s. At about the same time the prices of both types of assets surged in Finland, Norway, and Sweden. The third wave of bubbles initially was centered on Thailand and some of its neighbors in Southeast Asia. The fourth wave of bubbles primarily was in the real estate markets in the United States, Britain, Spain, Ireland, and Iceland.

Each of these waves of credit bubbles involved cross-border flows of money, which induced large increases in the values of the currencies and increases in the prices of real estate or stocks in the countries receiving the money.

Bubbles always implode, since by definition they involve non-sustainable increases in the indebtedness of a group of borrowers or non-sustainable increases in the prices of stocks. Debt can increase much more rapidly than income for two or three or a few more years, but debt cannot grow more rapidly than income for an extended period. When debt increases at 20 to 30 percent a year, the borrowers have an impeccable record for paying the scheduled interest in a timely way. Eventually the rate of growth of their indebtedness slows, and the ‘day of reckoning occurs’ when there isn’t enough cash from new loans to pay the interest on outstanding loans. Then the prices of real estate and of stocks decline. Moreover when the rate of growth of indebtedness slows, the currencies depreciate, and often very sharply.

When real estate prices decline the borrowers are the first group to incur losses; after they default, the losses cascade to the lenders. The implosion of the real estate and stock bubbles in Japan led to the massive failure of banks and a prolonged period of below-trend growth. The implosion of the asset price bubble in Thailand in mid-1997 triggered declines in currency values and asset prices throughout the region; recessions followed. However, there were no significant failures of US financial firms when the US stock prices declined by 40 percent between 2001 and 2003, and the ensuing recession was brief and shallow.

The range of movement in the values of national currencies since the early 1970s has been much larger than ever before. In 1971 the United States abandoned the US gold parity of $35 an ounce that had been established in 1934. The effort to retain a modified version of the Bretton Woods system of pegged currencies that was formalized in the Smithsonian Agreement of 1972 failed and a floating exchange rate arrangement was adopted by default early in 1973. At the beginning of the 1970s, the dominant market view was that the German mark and the Japanese yen might appreciate by 10 to 12 percent because their inflation rates had been below the US rate in the previous few years. The German mark and the Japanese yen appreciated more rapidly than anticipated through most of the 1970s, and then both currencies depreciated significantly in the first half of the 1980s, although not to the levels of the early 1970s. The Mexican peso, the Brazilian cruzeiro, the Argentinean peso and the currencies of many of the other developing countries depreciated by 30 to 40 percent or more in the early 1980s. The Finnish markka, the Swedish krona, the British pound, the Italian lira and the Spanish peseta lost more than a third of their value in the last six months of 1992. The Mexican peso lost more than half of its value during the presidential transition at the end of 1994. Most of the Asian currencies – the Thai baht, the Malaysian ringgit, the Indonesian rupiah and the South Korean won – depreciated sharply during the Asian Financial Crisis in the summer and autumn of 1997. The Argentinean peso lost more than two-thirds of its value in the first few months of 2001. The Icelandic krona lost half of its value in 2008. The euro, the new currency that eleven members of the European Union adopted at the beginning of 1999, soon depreciated by 30 percent, and then appreciated by 50 percent beginning in 2002.

The changes in the values of these individual currencies were much larger than those that would have been forecast based on the differences between the US and the foreign inflation rates. The ‘overshooting’ and ‘undershooting’ of national currencies were much larger than in any previous period.

The increases in commodity prices in the 1970s were spectacular. The US dollar price of gold increased from $40 an ounce at the beginning of the 1970s to nearly $1000 ten years later; the price was $450 at the end of the 1980s and $283 at the end of the 1990s. The price exceeded $1200 in the summer of 2010. The price of oil was $2.50 a barrel at the beginning of the 1970s and $40 at the end of that decade; in the mid-1980s the oil price was $12 and then at the end of the 1980s the price increased to $40 after the Iraqi invasion of Kuwait. The oil price almost reached $150 in the early summer of 2008, and then declined below $50 and then increased to $80.

The number of bank failures during the 1980s and the 1990s was much, much larger than in earlier decades. Several of these failures were isolated events: both Franklin National Bank in New York City and Herstatt AG in Cologne, Germany, had made large bets on the changes in currency values in the early 1970s that they subsequently lost. Crédit Lyonnais, once the largest bank in France and a government-owned firm, rapidly increased its loans in the effort to become a first-tier international bank and its bad loans eventually cost the French taxpayers the equivalent of more than $30 billion. However, most failures of banks and other financial firms were systemic and reflected dramatic changes in the financial environment. Three thousand US savings and loan associations and other thrift institutions failed in the 1980s, with losses to the American taxpayers of more than $100 billion.

When the bubbles in Japanese real estate and stocks imploded, the losses incurred by the Japanese banks were several times larger than their capital and virtually all the Japanese banks implicitly became wards of the government. Similarly when the Mexican peso and the currencies of the other developing countries depreciated sharply in the early 1980s, most of the banks in these countries went under because of the combination of the large loan losses by their domestic borrowers, in part due to the massive currency revaluation losses they had incurred. Virtually all of the banks in Finland, Norway, and Sweden went bankrupt when the bubbles in their real estate and stock markets imploded in the first half of the 1990s. Most of the Mexican banks failed at the end of 1994 when the peso depreciated sharply. Similarly most of the banks in Thailand and Malaysia and South Korea and several of the other Asian countries – except for Hong Kong and Singapore – tanked after the mid-1997 Asian Financial Crisis. The sharp declines in prices of residential real estate in the United States, Britain, Ireland, and several other countries that began toward the end of 2006 led to massive government investments – ‘bailouts’ – of the financial institutions. In 2008 many of the top firms in the US investment banking industry were wiped out or forced to seek a stronger merger partner. The British government ‘nationalized’ Northern Rock, the largest mortgage lender in the country, and became the dominant shareholder in the Royal Bank of Scotland. The Irish government made massive investments in the six largest banks in the country. The three large banks in Iceland were taken over by the government. Countrywide Financial, the largest mortgage lender in the United States, was acquired by Bank of America, which subsequently acquired Merrill Lynch, one of the largest US investment banks – but then Bank of America required a large injection of capital from the US Treasury. The US government made a massive investment in Citibank. The Dutch government provided capital to ING, the insurance–banking conglomerate.

These financial crises and bank failures resulted from the implosion of the asset price bubbles and from the sharp depreciations of currencies; in many cases the currency crises triggered the banking crises. The cost of these bank crises was extremely high in terms of several metrics – the losses incurred by the banks as a ratio of a country’s GDP and as a share of government spending, and the slowdowns in the rates of economic growth and the increases in unemployment and in the output gaps.

The massive number of bank failures, the large changes in currency values, and the asset price bubbles were systematically related and resulted from rapid changes in the global economic environment. The 1970s was a decade of accelerating inflation, the largest-ever sustained increase in the US price level in peace-time. The market price of gold surged because some investors relied on the cliché that ‘gold is a good inflation hedge’; however the increase in the gold price was many times larger than the contemporary increase in the US and world price levels. Toward the end of the 1970s investors were buying gold because its price was increasing – and the price was increasing because investors were buying gold.

The prevailing view in the late 1970s was that the US and world inflation rates would accelerate. Some analysts predicted that the gold price would reach $2500 an ounce and that the oil price would reach $80 to $90 a barrel by 1990.

The range of movement in bond prices and stock prices in the 1970s was much greater than in the several previous decades. In the 1970s the real rates of return on both US dollar bonds and US stocks were negative. In contrast in the 1990s the real rates of return on bonds and on stocks averaged more than 15 percent a year.

The foreign indebtedness of Mexico, Brazil, Argentina, and other developing countries as a group increased from $125 billion in 1972 to $800 billion in 1982. One cliché at the time was that ‘countries don’t go bankrupt’. During this period the borrowers had a stellar record for paying the interest on their loans on a timely basis. Then in the autumn of 1979 the Federal Reserve adopted a sharply contractive monetary policy; interest rates on US dollar securities surged. The price of gold peaked in January 1980 and then began to decline as inflationary anticipations were reversed.

The sharp increase in real estate prices and stock prices in Japan in the 1980s was associated with a boom in the economy; Japan as Number One: Lessons for America¹ was a bestseller in Tokyo. The Japanese banks increased their deposits and their loans and their capital much more rapidly than banks headquartered in the United States and in Germany and in the other European countries. At the time seven or eight of the ten largest banks in the world were Japanese. Then at the beginning of the 1990s real estate prices and stock prices in Japan imploded. Within a few years many of the leading Japanese banks and financial institutions were broke, kaput, bankrupt, and insolvent, and remained in business only because of an implicit understanding that the Japanese government would protect the depositors from financial losses if the banks were closed. A striking story of a mania and a crash – but without a panic, because depositors believed that government would socialize the loan losses.

Three of the Nordic countries – Norway, Sweden, and Finland – experienced bubbles in their stock markets and real estate markets at about the same time as a result of money inflows associated with financial liberalization. Their bubbles popped at about the same time as the one in Japan.

Mexico had been one of the great economic success stories of the early 1990s as it prepared to enter the North American Free Trade Agreement. The Bank of Mexico had adopted a tough contractive monetary policy that reduced the inflation rate from 140 percent to less than 10 percent in four years; during the same period several hundred government-owned firms were privatized and business regulations were liberalized. Money flowed to Mexico because the real rates of return on government securities were high and because the prospective profit rates on industrial investments were also high. The universal expectation was that Mexico would become the low-cost base for producing automobiles and washing machines and many other manufactured goods for the US and Canadian markets. The large money inflow led to a real appreciation of the peso, Mexico’s trade deficit increased to 7 percent of its GDP and its external debt to 60 percent of its GDP. Then several political incidents associated with the presidential election in 1994 led to a sharp decline in the flow of money and the Mexican government was unable to support the peso. Once again the depreciation of the peso resulted in large loan losses, and most of the Mexican banks – which had been privatized in the previous several years – failed.

In the mid-1990s real estate prices and stock prices surged in Thailand, Malaysia, and Indonesia; these were the ‘dragon economies’ that seemed likely to emulate the economic successes of the ‘Asian tigers’ – Taiwan, South Korea, Hong Kong, and Singapore – of the previous generation . Firms based in Japan, Europe, and the United States invested in these countries as low-cost sources of supply, much as US and foreign firms had invested in Mexico as a source of supply for the North American market. European and Japanese banks rapidly increased their loans to firms and banks in these countries. The domestic lenders in Thailand then experienced large losses on their domestic loans in the autumn and winter of 1996 because they had not been sufficiently discriminating in their evaluations of the willingness of Thai borrowers to pay the interest on their indebtedness. Foreign lenders sharply reduced their purchases of Thai securities, and then the Bank of Thailand, much like the Bank of Mexico thirty months earlier, did not have the money to support its currency. The sharp decline in the value of the baht in early July 1997 led to money outflows from the other Asian countries and their currencies (except for the Hong Kong dollar and the Chinese yuan, which remained rigidly pegged to the US dollar) declined by 30 percent or more. The Indonesian rupiah lost 80 percent of its value. Most of the banks in the area – except for those in Hong Kong and Singapore – would have been bankrupt in any reasonable ‘mark-to-market’ test. The crisis spread to Russia, there was a debacle in the ruble, and the country’s banking system collapsed in the summer of 1998. Investors then became more cautious and they sold risky securities and bought safer US government securities, and the changes in the relationship between the interest rates on these two groups of securities led to the collapse of Long-Term Capital Management, then the largest US hedge fund.

In part the large number of crashes in national financial markets in the last thirty years reflects that there are more independent countries. Despite the lack of perfect comparability across periods, the conclusion is unmistakable that financial failure has been more extensive and pervasive in the last thirty years.

The bubble in US stock prices in the second half of the 1990s was associated with a remarkable US economic boom; the unemployment rate declined sharply, the inflation rate declined, and the rates of economic growth and productivity both accelerated. The US government developed its largest-ever fiscal surplus in 2000 after having had its largest-ever fiscal deficit in 1990. The remarkable performance of the real economy contributed to the surge in US stock prices that in turn led to the increase in investment spending and consumption spending and an increase in the rate of US economic growth.

US stock prices began to decline in the spring of 2000 and fell by 40 percent in the next three years while the prices of NASDAQ stocks declined by 80 percent.

US real estate prices began to increase at an above-average rate in 2002. Real estate prices increase in the long run, in part because of the increase in the general price level and in part because of the increase in nominal GDP. (Much of the increase in real estate prices reflects increases in the price of land.) The Federal Reserve maintained low interest rates in part because of the sluggishness in the economy, and house prices increased three times as rapidly as the general price level. The sharp increase in prices induced a construction boom, and housing starts reached two million units a year – about 500,000 more units than the number required to satisfy the growth in population and the losses to fires, storms, and similar factors. Part of the increase in demand was from investors who sought profits from the continued increases in prices.

The sharp decline in the price of residential real estate and the debacle in the prices of mortgage-related securities since 2007 has led to many books that are US-centric, and that seek to explain the bubble in terms of the failure of regulation or the greed of the bankers or the failure of the regulators or vagaries of new financial instruments.

One of the themes of this book is that the credit bubbles that often occur in several different countries at the same time have similar initial causes. Thus the surge in the indebtedness of the developing countries in the 1970s occurred because the major international banks believed that commodity prices would continue to increase and that the growth rates in these countries would remain high. The likelihood that the bubbles in the real estate markets in the United States, Britain, Ireland, Iceland, Spain, South Africa, and several other countries that began about 2002 were independent events seems low. Bubbles in real estate always result from bubbles in the growth of credit. There were unique idiosyncratic aspects in these different national markets; the market in subprime mortgages seems uniquely American. The rapid growth in the supply of credit led to a sharp increase in the demand for mortgages, which was greater than the supply of prime loans and so the mortgage brokers ginned up a large increase in the supply of sub-prime mortgages.

Another theme is that the likelihood that the four waves of bubbles over a thirty-year period were unrelated events is low. Each bubble led to a crisis. Several of these crises appear to have laid the basis for the next wave of bubbles. The financial crisis in the developing countries in the early 1980s had a knock-on effect that contributed to the bubble in Japanese real estate and stocks in the second half of the 1980s. The implosion of the bubble in Tokyo in the early 1990s led to an increase in money flows from Japan to Thailand and Malaysia and Indonesia, which led to the appreciation of their currencies and to increases in the prices of real estate and of securities in these countries. When the bubbles in the countries in Southeast Asia imploded, there was surge in the flow of money to the United States as these countries repaid loans; the US dollar appreciated and the US trade deficit increased by $150 billion a year.

The increase in the flow of money to a country almost always led to increases in value of its currency and to increases in the prices of assets in that country as the domestic sellers of the securities used nearly all of their receipts to buy other securities from other domestic residents. These domestic residents in turn similarly used a large part of their receipts to buy other domestic securities from other domestic residents. These transactions in securities occurred at ever-higher prices. It was as if the cash from the sale of securities to foreigners was the proverbial ‘hot potato’ that was rapidly passed from one group of investors to others at ever-increasing prices.

Manias and credit and books

The production of books on financial crises is counter-cyclical. A spate of books on the topic appeared in the 1930s following the US stock market bubble in the late 1920s and the subsequent crash and the Great Depression. Relatively few books on crises appeared during the several decades immediately after World War II.

The first edition of this book was published in 1978, after US stock prices had declined by 50 percent in 1973 and 1974 following a fifteen-year bull market in stocks. The stock market debacle and the US recession led to the bankruptcies of the Penn Central railroad, several of the large steel companies and a large number of Wall Street brokerage firms. New York City was on the verge of default on its outstanding bonds and was saved from insolvency by the State of New York. Not quite a crash, unless you were a senior official or a stockholder in one of the firms that failed or the Mayor of New York City.

The fifth edition was published after the implosion of the dot.com bubble in US stocks in the late 1990s. The innovation since the fifth edition were the bubble in residential real estate in the United States, Britain, Ireland, Spain, Iceland, and several other countries, and the sharp increase in the indebtedness of the governments of Greece, Portugal, Ireland, and Spain.

Each of these waves of bubbles has been global, in that four, five, or more countries have been involved. Moreover the bubbles seem larger, judged by the increase in household wealth.

The sixth edition of this book appears thirty years after the first and also after thirty of the most tumultuous years in global financial markets, a period without a good historical precedent in terms of monetary turbulence.

The earliest bubble noted in the box involved tulip bulbs in the Netherlands in the seventeenth century. Two of the bubbles – one in Britain and one in France – occurred at the end of the Napoleonic Wars. There were manias and financial crises in the nineteenth century that were mostly associated with the failures of banks, often after an extended investment in infrastructure such as canals and railroads. Currency crises and banking crises were frequent between 1920 and 1940. The percentage increases in stock prices in the past thirty years have been larger than in earlier periods, and six of the ten bubbles in the box have occurred in this period. Bubbles in real estate and in stocks have often occurred together; some countries have experienced a bubble in real estate but not in stocks, while the United States had a stock price bubble in the second half of the 1990s that had no counterpart in the real estate market.

Manias are dramatic but they have been infrequent; only two have occurred in US stocks in two hundred years. Manias generally have occurred during the expansion phase of the business cycle, in part because the euphoria associated with the mania leads to increases in spending. During the mania the increases in the prices of real estate or stocks or in one or several commodities contribute to increases in consumption and investment spending that in turn lead to quickening of economic growth. Seers in the economy forecast perpetual economic growth and some venturesome ones even proclaim an end to recessions and declare that traditional business cycles have become obsolete. The more rapid growth induces investors and lenders to become more optimistic, and asset prices increase more rapidly.

Manias – especially macro manias – are associated with economic euphoria; business firms become increasingly upbeat and investment spending surges because credit is plentiful. In the second half of the 1980s Japanese industrial firms could borrow as much as they wanted from their friendly bankers in Tokyo and in Osaka; money seemed ‘free’ (money always seems free in manias) and the Japanese went on both a consumption spree and an investment spree. The Japanese purchased ten thousand items of French art. A racetrack entrepreneur from Osaka paid $90 million for Van Gogh’s Portrait of Dr Guichet, at that time the highest price ever paid for a painting. The Mitsui Real Estate Company paid $625 million for the Exxon Building in New York even though the initial asking price had been $510 million; Mitsui wanted to get in the Guinness Book of World Records for paying the highest-ever price for an office building. In the second half of the 1990s in the United States newly established firms in the information technology industry and in biotech had access to virtually unlimited funds from the venture capitalists who believed they would profit greatly when the shares in these firms were first sold to the public.

During these euphoric periods an increasing number of investors seek short-term capital gains from the increases in the prices of real estate and of stocks. Investors make down payments for the purchase of condominium apartments in the pre-construction phase in the anticipation that they will be able to sell these apartments at handsome profits when the buildings have been completed, or even before.

Then an event – perhaps a change in government policy, an unexplained failure of a firm previously thought to have been successful – occurs that leads to a pause in the increase in asset prices. Soon, some of the investors who had financed most of their purchases with borrowed money become distress sellers because the interest payments on the money borrowed to finance their purchases are larger than the investment income on the assets. The prices of these assets decline below their purchase prices and now the buyers are ‘under water’ – the amounts owed on the money borrowed to finance the purchases of these assets are larger than their current market value. Their distress sales lead to sharp declines in the prices of the assets and a crash and panic are likely to follow.

The economic situation in a country after several years of bubble-like behavior resembles that of a young person on a bicycle; the rider needs to maintain the forward momentum or the bike becomes unstable. During the mania, asset prices will decline immediately after they stop increasing – there is no plateau, no ‘middle ground’. The decline in the prices of some assets leads to the concern that asset prices will decline further and that the financial system will experience ‘distress’. The rush to sell these assets becomes self-fulfilling and so precipitous that it resembles a panic. The prices of commodities – houses, buildings, land, stocks, bonds – crash to levels that are just 30 to 40 percent of their prices at the peak. Bankruptcies surge, economic activity slows, and unemployment increases.

The features of these manias are never identical and yet there is a similar pattern. The increase in prices of commodities or real estate or stocks is associated with euphoria; household wealth increases and so does spending. There is a sense of ‘We never had it so good’. Then the asset prices peak and then begin to decline. The implosion of a bubble leads to declines in the prices of commodities, stocks and real estate. Some financial crises were preceded by a rapid increase in the indebtedness of one or several groups of borrowers rather than by a rapid increase in the price of an asset or a security.

The thesis of this book is that the cycle of manias and panics results from the pro-cyclical changes in the supply of credit; the credit supply increases rapidly in good times, and then when economic growth slackens, the rate of growth of credit has often declined sharply. A mania involves increases in the prices of real estate or stocks or a currency or a commodity in the present and near-future that are not consistent with the prices of the same real estate or stocks in the distant future. The forecasts that the price of oil would increase to $80 a barrel after the earlier increase from $2.50 a barrel at the beginning of the 1970s to $36 at the end of that decade was manic. During the economic expansions investors become increasingly optimistic and more eager to pursue profit opportunities that will pay off in the distant future while the lenders become less risk-averse. Rational exuberance morphs into irrational exuberance, economic euphoria develops and investment spending and consumption spending increase. There is a pervasive sense that it is ‘time to get on the train before it leaves the station’ and the exceptionally profitable opportunities disappear. Asset prices increase further. An increasingly large share of the purchases of these assets is undertaken in anticipation of short-term capital gains and an exceptionally large share of these purchases is financed with credit.

The financial crises that are analyzed in this book are major both in size and in effect and most are international because they involve several different countries either at the same time or in a causal, sequential way.

The term ‘bubble’ is a generic term for the increases in asset prices in the mania phase of the cycle that cannot be explained by the changes in the economic fundamentals. Recently, real estate bubbles and stock price bubbles have occurred at more or less the same time in Japan and in some of the Asian countries. The sharp increases in the prices of gold and silver in the late 1970s were a bubble, but the increases in the price of crude petroleum in the same years were not; the distinction is that many of the buyers of gold and silver in that tumultuous and inflationary decade anticipated that the prices of both precious metals would continue to increase and that profits could be made from buying and holding these commodities for relatively short periods. In contrast many of the buyers of petroleum were concerned that the disruptions in oil supplies due to actions of the cartel and the war in the Persian Gulf would lead to shortages and increases in prices.

Virtually every mania is associated with a robust economic expansion, but only a few economic expansions are associated with a mania. Still, the association between manias and economic expansions is sufficiently frequent and sufficiently uniform to merit renewed study.

Some economists have contested the view that the use of the term bubble is appropriate because it suggests irrational behavior that is highly unlikely or implausible; instead they seek to explain the rapid increase in real estate prices or stock prices in terms that are consistent with changes in the economic fundamentals. Thus, for them, the surge in the prices of NASDAQ stocks in the 1990s occurred because investors sought to buy shares in firms that would repeat the spectacular successes of Microsoft, Intel, Cisco, Dell, and Amgen.

The policy implications

The appearance of a mania or a bubble raises the policy issue of whether governments should seek to moderate the surge in asset prices to reduce the likelihood or the severity of the ensuing financial crisis. Virtually every large country has established a central bank as a domestic ‘lender of last resort’ to reduce the likelihood that a shortage of liquidity would cascade into a solvency crisis. The practice leads to the question of the role for an international ‘lender of last resort’ that would assist countries in stabilizing the value of their currencies and reduce the likelihood that a sharp depreciation of the currencies because of a shortage of liquidity would trigger large numbers of bankruptcies.

During a crisis, many firms that had recently appeared robust tumble into bankruptcy because the failure of some firms often leads to a decline in asset prices and a slowdown in the economy. When asset prices decline sharply, government intervention may be desirable in order to provide the public good of stability. During financial crises the decline in asset prices may be so large and abrupt that the price changes become self-justifying. When asset prices tumble sharply, the surge in the demand for liquidity may drive many individuals and firms into bankruptcy, and the sale of assets in these distressed circumstances may induce further declines in their prices. At such times a lender of last resort can provide financial stability or attenuate financial instability. The dilemma is that if investors knew in advance that governmental support would be forthcoming under generous dispensations when asset prices fall sharply, markets might break down somewhat more frequently because investors will be less cautious in their purchases of assets and of securities.

The role of the lender of last resort in coping with a crash or panic is fraught with ambiguity and dilemma. Thomas Joplin commented on the behavior of the Bank of England in the crisis of 1825: ‘There are times when rules and precedents cannot be broken; others, when they cannot be adhered to with safety.’ Breaking the rule establishes a precedent and a new rule that can be adhered to or broken as occasion demands. In these circumstances intervention is an art rather than a science. The general rules that the state should always intervene or that the state should never intervene are both wrong. This same issue of intervention reappeared with the question of whether the US government should have rescued Chrysler in 1979, New York City in 1975 and the Continental Illinois Bank in 1984. (In fact Continental Illinois failed, although the depositors in the bank were made whole.). Similarly, should the Bank of England have rescued Baring Brothers in 1995 after the rogue trader Nick Leeson in its Singapore branch office had depleted the firm’s capital through hidden transactions in option contracts? The question appears whenever a group of borrowers or banks or other financial institutions incurs such massive losses that they are likely to be forced to close, at least under their current owners. The United States acted as the lender of last resort during the Mexican financial crisis at the end of 1994. The International Monetary Fund acted as the lender of last resort during the Russian financial crisis of 1998, primarily after prodding by the US and German governments. Neither the United States nor the International Monetary Fund was willing to act as a lender of last resort during the Argentinean financial crisis at the beginning of 2001. This list highlights that coping with financial crises remains a major problem.

The conclusion of The World in Depression, 1929–1939 was that the 1930s depression was wide, deep and prolonged because there was no international lender of last resort.² Britain was unable to act in that capacity because it was exhausted by World War I, obsessed with pegging the British pound to gold at its pre-1914 parity and groggy from the aborted economic recovery of the 1920s. The United States was unwilling to act as an international lender of last resort; at the time few Americans had thought through what the United States might have done in that role. This book extends the analysis of the responsibilities of an international lender of last resort (Chapter 12).

The monetary aspects of manias and panics are important and are examined at length in several chapters. The monetarist view – at least one monetarist view – is that the mania would not occur if the rate of growth of the money supply were stabilized or constant. Many of the manias are associated with the surge in the growth of credit, but some are not; a constant money supply growth rate might reduce the frequency of manias but is unlikely to consign them to the dustbins of history. The rate of increase in US stock prices in the second half of the 1920s was exceptionally high relative to the rate of growth of the money supply, and similarly the rate of increase in the prices of NASDAQ stocks in the second half of the 1990s was exceedingly high relative to the growth of the US money supply. Some monetarists distinguish between ‘real’ financial crises that are caused by the shrinkage of the monetary base or high-powered money and ‘pseudo’ crises that do not. The financial crises in which the monetary base changes early or late in the process should be distinguished from those in which the money supply did not increase significantly.

The earliest manias discussed in the first edition of this book were the South Sea and Mississippi bubbles of 1719–20. The earliest manias analyzed in this edition are the Kipper- und Wipperzeit, a monetary crisis (1619–22) that occurred at the outbreak of the Thirty Years War, and the much-discussed ‘tulipmania’ of 1636–37. The view that the trade in tulip bulbs in the Dutch Republic constituted a bubble followed from widespread recognition, even at the time, that exotic specimens of tulips are difficult to breed, but once bred propagate easily – and hence eventually their prices would decline sharply.³

The early-historical treatment focusses on European experiences. The most recent crisis covered in this edition centers on the real estate markets in the United States, Britain, Ireland, Spain, and Iceland. The concentration on the financial crises in Britain in the nineteenth century reflects both the central importance of London in international financial arrangements and the abundant writings by contemporary analysts. In contrast, Amsterdam was the dominant financial power for much of the eighteenth century, but events there are rather glossed over because of the difficulties in accessing the Dutch literature.

The waves of credit bubbles and crises since the mid-1970s suggest that these market events were much more global then in the past. Most of the countries that have experienced credit bubbles also have received an inflow of money. Because few currencies are pegged, the inflows have led to an appreciation of their currencies and – since the money has to go somewhere – to increases in asset prices. Each of these countries has experienced an economic boom – ‘the times could not be better’. Perhaps because the currencies have appreciated, upward pressures on prices of goods and services have been smaller than they would have been had the currencies been pegged. Nevertheless the central banks in many countries raised interest rates – which had the impact of attracting more money from abroad.

One unique feature of these monetary developments is that the links among the global markets means that money is more likely to move abroad for a smaller differential in anticipated returns. The innovation is that there is a larger pool of liquid funds denominated in the US dollar that investors can tap when they want to buy assets and securities in countries whose financial and economic prospects suddenly look much brighter.

The story chapter by chapter

A stylized model of speculation, credit expansion, financial distress, and then crisis that ends in a panic and crash is presented in Chapter 2. The model follows the early classical ideas of ‘over-trading’ followed by ‘revulsion’ and ‘discredit’ – musty terms used by earlier generations of economists including Adam Smith, John Stuart Mill, Knut Wicksell and Irving Fisher. These concepts were developed further by Hyman Minsky, who argued that the financial system in a market economy is unstable, fragile, and prone to crisis. The Minsky model has great explanatory power for earlier crises in the United States and in Western Europe, for the asset price bubbles in Japan in the second half of the 1980s, and for the bubbles in real estate in the United States, Britain, Ireland, Spain, and Iceland between 2002 and 2007.

The mania phase of the economic expansion is the subject of Chapter 3. The central issue is whether markets are always rational, or whether speculation can be destabilizing – do investors in real estate and in stocks develop estimates of the anticipated prices on the basis of recent increases in the prices of these assets or are their anticipations of prices based on their estimates of their earning power. The nature of the outside, exogenous shock that triggers the mania is examined in different historical settings including the onset and the end of a war, a series of good harvests and a series of bad harvests, the opening of new markets and of new sources of supply and the development of different innovations – the railroad, electricity and e-mail. A particular recent form of displacement that shocks the system has been financial liberalization or deregulation in Japan, the Nordic countries, some of the Asian countries, Mexico, Russia, and Iceland. Deregulation has led to monetary expansion, foreign borrowing and speculative investment.⁴

Investors have speculated in commodities, agricultural land, urban building sites, railroads, new banks, discount houses, stocks, bonds (both foreign and domestic), glamour stocks, conglomerates, condominiums, shopping centers and office buildings. Moderate excesses burn themselves out without damage to the economy although individual investors encounter large losses. One question is whether the euphoria of the economic upswing endangers financial stability only if it involves at least two or more objects of speculation, a bad harvest, say, along with a railroad mania or an orgy of land speculation, or a bubble in real estate and in stocks at the same time.

The monetary dimensions of both manias and panics are analyzed in Chapter 4. The occasions when a boom or a panic has been triggered by a monetary event – a re-coinage, a discovery of precious metals, a change in the ratio of the prices of gold and silver under bimetallism, an unexpected success of some flotation of a stock or bond, a sharp reduction in interest rates as a result of a massive debt conversion, or a rapid expansion of the monetary base – are noted. Innovations in finance, as in productive processes, can shock the system and lead to overinvestment in some types of financial services.⁵ The financial sectors expanded rapidly in the United States, Britain, Ireland, and Iceland during their real estate booms because of the rapid increases in mortgages and mortgage-related securities. A sharp increase in interest rates may induce some investors to withdraw money from banks and thrift institutions, which are then squeezed because the prices of their long-term securities decline when they need to downsize.

The problems associated with managing the monetary mechanism to avoid manias and bubbles is stressed in this edition. Most bubbles, and especially those since the 1970s, have resulted from the rapid increase in the credit available to a group of borrowers, often the buyers of real estate. Monetary control by central banks limits the growth of money and credit. Money is a public good but monetary arrangements often have been exploited by private parties. Banking, moreover, is difficult to regulate because new institutions are developed that circumvent the regulations. Many monetarists insist that many, perhaps most, of the cyclical difficulties of the past have resulted from mismanagement of the monetary mechanism. Such mistakes were frequent and serious. The argument advanced in Chapter 4, however, is that even when the supply of money was adjusted to the demands of an economy the monetary mechanism did not stay right for very long. When government produces one quantity of the public good, money, the public may proceed to produce many close substitutes for money, just as lawyers find new loopholes in tax laws almost as fast as older ones are closed. The evolution of money from coins to bank notes, bills of exchange, bank deposits and finance paper illustrates the point. The Currency School may have been right about the need for a fixed supply of money, but it was wrong to believe that the money supply could be fixed forever.

The domestic aspects of the crisis are reviewed in Chapter 5. One question is whether manias can be halted by official warnings – moral suasion and jawboning. The evidence suggests that they cannot, or at least that many crises followed warnings that were intended to head them off. One widely noted remark was that of Alan Greenspan, chairman of the Federal Reserve Board, who stated on 6 December 1996 that he thought that the US stock market was ‘irrationally exuberant’. The Dow Jones industrial average was 6600; subsequently the Dow peaked at 11,700. The NASDAQ had been at 1300 at the time of the Greenspan remark and peaked at more than 5000 four years later. A similar warning had been issued in February 1929 by Paul M. Warburg, a private banker who was one of the fathers of the Federal Reserve system, without slowing the stock market’s upward climb. The nature of the event that ultimately produces a turning point is discussed: some bankruptcy, defalcation or troubled area revealed or rumored, or a sharp rise in the central bank discount rate to halt the hemorrhage of cash into domestic circulation or abroad. And then there is the interaction of falling prices – a crash – and its impact on the liquidity in the economy.

The impacts of the mania on domestic spending and the resulting euphoria are discussed in Chapter 6. Bubbles lead to extravagant expenditure. Malaysia, Dubai, and a few other places have all built the tallest building in the world – to show that they could do it and afford to pay for it. The Japanese imported French art in the 1980s – because other Japanese were importing French art. Money seems ‘free’ as if the fundamental laws of economics no longer apply. But observations of extravagant expenditures eventually lead to questions of whether there is an underlying bubble.

The implosion of a bubble always leads to the discovery of frauds and swindles that developed in the froth of the mania; these events are reviewed in Chapter 7. Fraud and corruption are based on mis-information – both falsification and misrepresentation; some fraud also involves the theft of private information before it becomes publicly available. Some of the fraud is personal, some is corporate. Bernie Madoff ran one of the largest Ponzi schemes ever, investors lost more than $20 billion. The owners of some business conglomerates in Iceland had ‘captured’ control of the banks and then borrowed from the banks to increase their consumption and their investments. The combination of failed thrift institutions and the rapid growth of junk bonds in the 1980s cost American taxpayers more than $100 billion; some of these thrifts had been acquired by individuals who relied on the junk bonds for their financing. Enron, MCIWorldCom, Tyco, Dynegy, and Adelphia were a rogue’s gallery of the 1990s. Many of the large US mutual fund families were exposed because they provide favored treatment to hedge funds. Crashes and panics are often precipitated by the revelation of some misfeasance, malfeasance or malversation (the corruption of officials) that occurred during the mania. One inference is that the swindles are a response to the appetite for wealth (or plain greed) stimulated by the boom; the Smiths want to keep up with the Joneses and some Smiths engage in fraudulent behavior in order to do so. As the monetary system gets stretched, institutions lose liquidity and as unsuccessful swindles seem about to be revealed, the temptation to take the money and run becomes irresistible.

The international contagion of manias and crises from the seventeenth to the first half of the twentieth century is the subject of Chapter 8; two, three, four, or more countries experienced similar bubble symptoms at the same time. There are two competing narratives. One involves the metaphor of the sun and the moon, with the mania initiated in one country, often a large one, and then ‘beamed’ to its smaller neighbors or trading partners. Its alternative posits that these countries are subject to the same shock or innovation. There are many possible linkages among countries, including trade, arbitrage of securities, capital flows, changes in central bank reserves of gold or other international reserve assets, and direct contagion of speculators in euphoria or gloom. Some bubbles are national, others global. Some crises are national, others international. What constitutes the difference? Did, for example, the 1907 panic in New York precipitate the collapse of the Società Bancaria ltaliana via pressure on Paris communicated to Turin by withdrawals of bank deposits? There is a fundamental ambiguity – tight money in one financial center can serve either to attract funds or to repel them, depending on the expectations that a rise in interest rates generates. With inelastic expectations – no fear of crisis or of currency depreciation – an increase in the discount rate attracts funds from abroad and helps provide the cash needed to enhance liquidity; with elastic expectations of changes – of falling prices, bankruptcies, or currency depreciation – raising the discount rate may suggest the need to take more funds out rather than bring new funds in. The dilemma is familiar in economic life generally. A rise in the price of a commodity may lead consumers to postpone purchases in anticipation of a decline, or to speed up purchases before prices rise further. And even where expectations are inelastic, and the increase in the discount rate at the central bank sets in motion the right reactions, lags in responses may be so long that the crisis begins before the Marines arrive.

One complex but not unusual trigger that leads to financial crisis is a sudden halt to foreign lending, perhaps because of a domestic boom; thus the boom in Germany and Austria in 1873 led to a decline in money outflows and contributed to the difficulties of Jay Cooke in the United States. Similar developments occurred with the Baring crisis in 1890, when troubles in Argentina led to a sudden decline in money flows to South Africa, Australia, the United States and other Latin American countries. The stock market boom in New York in the late 1920s led Americans to buy fewer of the new bond issues of Germany and various Latin American countries, which in turn caused these countries to slide into depression. A halt to foreign trading is likely to precipitate depression abroad, which may in turn feed back to the country that launched the process.⁶

The discussion in Chapter 9 highlights the four waves of credit bubbles since the mid-1970s, and the relationships among the successive waves. The likelihood that these four waves are independent and unrelated seems low. The first of these waves involved the surge in bank loans to governments and government-owned firms in the larger developing countries in the 1970s, and the second bubble was in Tokyo in the second half of the 1980s; was there a connection between the developing country debt crisis and the bubble in Japan? The third wave was in Thailand, Malaysia, Indonesia, Russia, and other developing countries in the mid-1990s, while the fourth wave occurred in the real estate markets in the United States, Britain, Ireland, Iceland, and Spain ten years later. The crisis in the debt of the governments of Greece and other Mediterranean countries may be a follow-on to the fourth wave. Asset bubbles would not occur without the rapid growth of credit, either in the aggregate or to particular groups of borrowers. The theme of this chapter is that there was a systematic relationship among these waves, or at least between the first and the second, and the second and the third. The credit bubble that centered on Mexico and the other developing countries was sui generis. When the bubble in bank loans to the developing countries imploded in 1982, the currencies of these countries depreciated sharply – but the Bank of Japan resisted the pressure toward the appreciation of the yen in the second half of the 1980s, which led to a rapid increase in the money supply in Japan. When the bubble in Japan imploded at the beginning of the 1990s, the yen appreciated, and Japanese firms rapidly increased their investments in productive facilities in Southeast Asia. The currencies of these countries appreciated, and property prices increased. When the bubble in stock prices and real estate prices in Bangkok and the other Asian capitals imploded in 1997 and 1998, there was a surge in the flow of money to New York as Asian borrowers repaid their foreign loans; the Asian currencies depreciated sharply.

Domestic crisis management is reviewed in Chapter 10 and 11. The first of these two chapters considers the range of domestic responses to a crisis; at one extreme the government may take a hands-off position, at the other there is a range of miscellaneous measures. Those who believe that the market is rational often prefer the hands-off approach; according to one formulation, it is healthy for the economy to go through the purgative fires of deflation and bankruptcy to get rid of the mistakes and excesses of the boom. Among the miscellaneous devices are holidays, bank holidays, the issue of scrip, guarantees of liabilities, issuance of government debt, deposit insurance and the formation of special institutions like the Reconstruction Finance Corporation in the United States in 1932 and the Istituto per la Ricostruzione Industriale (IRI) in Italy in 1933. The Italian literature calls the process the ‘salvage’ of banks and companies; the British in 1974–75 referred to saving the fringe banks as a ‘lifeboat’ operation.

Issues related to a domestic lender of last resort are the focus of Chapter 11 – primarily whether there should be such a lender, who this lender should be and how it should operate. A key topic is ‘moral hazard’ – if investors are confident that they will be ‘bailed out’ by a lender of last resort, their self-reliance may be weakened. But on the other hand, the priority may be to stop the panic, to ‘save the system today’ despite the adverse effects on the incentives of investors. If there is a lender of last resort, however, whom should it save? Insiders? Outsiders and insiders? Only the solvent, if illiquid? The distinction between solvent and insolvent firms depends on the scope and duration of the panic. These are political questions, and they are raised in particular when it becomes necessary to legislate to increase the capital of the Federal Deposit Insurance Corporation (FDIC) or the Federal Savings and Loan Insurance Corporation (FSLIC) when one or the other runs out of funds to bail out failed banks. The issue was particularly acute in the 1990s in Japan, where the collapse of the Nikkei stock bubble in 1990 uncovered all sorts of bad real estate loans by banks, credit unions and other financial houses, confronting the government with the headache of deciding how much of a burden to put on the taxpayer. If the taxpayers didn’t take the hit, then the owners of bonds and stocks would incur very large losses because the prices of these securities would decline further. Particularly troubling in this context was the catatonic state of government in Japan in the 1990s – slow to decide how to meet the crisis and even slower to act.

The role of an international lender of last resort in providing global monetary stability is the focus of Chapter 12. The problem is that there is no responsible government or agency of government with the de jure responsibility for providing this public good. US government support for Mexico, first in 1982 and again in 1994 was justified on the grounds that countries of the North American Free Trade Agreement (NAFTA) should stick together and that assistance to Mexico would dampen or neutralize the contagion effect and prevent a collapse of lending to Brazil and Argentina and other developing countries. The sharp depreciation of the Thai baht in the early summer of 1997 triggered crises in nearby Asian countries, including Indonesia, Malaysia, and South Korea as well as in Singapore, Hong Kong and Taiwan.

The thrust of Chapter 13 – a new chapter for the sixth edition – is the severity of the credit crisis that began in September 2008 – after the US government decided not to provide the financial assistance that would have allowed Lehman Brothers to remain a viable concern. In February the US Treasury and the Federal Reserve made an arrangement for JPMorgan Chase by agreeing that Morgan could put up to $29 billion of ‘toxic securities’ with the Fed. Shareholders of Bear Stearns received $10 a share – many of the shareholders were employees of Bear and this price was viewed as a retention bonus. During the first week of September Fannie Mae and Freddie Mac, the US government-sponsored lenders that accounted for more than 50 percent of the credit risk on home mortgages, were taken over by the US Treasury – both institutions were bankrupt and the owners of their common shares and their preferred shares lost virtually all of their money. It appeared as if the US government had in effect adopted a ‘too big to fail’ policy – these institutions would continue in business, although there would be a dramatic change in ownership. At the outset of the crisis, the markets viewed the decision not to ‘save Lehman Brothers’ as a major change in policy, but a day after Lehman closed its doors, there was a run on AIG, then the largest insurance company in the world, and the ‘too big to fail’ policy was resurrected. Nevertheless, investors still panicked, evidenced by the sharp increase in the spreads on riskier assets.

The final chapter seeks to answer two questions; the first is why there has been so much economic turmoil in the international financial economy in the last thirty years, and the second is whether an international lender of last resort would have made a difference. The International Monetary Fund was established in the 1940s to act as an international lender of last resort and to fill an institutional vacuum; the view was that financial crises in the 1920s and the 1930s would have been less severe had there been an international lender of last resort. The large number of crises in the last thirty years leads to the question of whether the presence of the IMF as a supplier of national currencies to countries with financial crises encouraged profligate national financial policies. Financial arrangements need a lender of last resort to prevent the escalation of the panics that are associated with crashes in asset prices. But a commitment to the view that a lender is needed should be distinguished from the view that individual borrowers will be ‘bailed out’ if they become over-extended. For example, uncertainty about whether New York City would be helped, and by whom, may have been the appropriate policy in the long run, so long as help was finally provided, and so long as there was doubt right to the end as to whether it would be forthcoming. This is a neat trick: always come to the rescue, in order to prevent needless deflation, but always leave it uncertain whether rescue will arrive in time or at all, so as to instill caution in other speculators, banks, cities or countries. The economic implosion triggered by the failure of Lehman Brothers could have been avoided if the firm had been acquired by another private firm much as Bear Stearns or if the government had taken over ownership of the firm. In Voltaire’s Candide, the head of a general was cut off ‘to encourage the others’. A sleight of hand may be necessary to ‘encourage’ the others (without, of course, cutting off actual heads) to participate in lender-of-last-resort activities because the alternative is likely to have very expensive consequences for the economic system.

Is the fourth wave of asset price bubbles since the mid-1970s the last in this series, or instead is it likely that the crisis that followed the collapse of real estate prices and the measures adopted to dampen the hardships associated with the failure of many banks in various countries have laid the foundation for a fifth wave. An epilogue summarizes some of the proposals for reform of the US financial system and asks whether the bubble and crisis of 2002–08 would have been significantly different if the legislation adopted in 2010 had been adopted in 2000.