2

The Anatomy of a Typical Crisis

History vs economics

Historians view each event as unique. In contrast economists search for the patterns in the data, and the systematic relationships between an event and its antecedents. History is particular; economics is general. The business cycle is a standard feature of market economies; increases in investment spending lead to increases in household income and in GDP growth. Macroeconomics focuses on the explanations for the cyclical variations in the growth of GDP relative to its long-run trend.

An economic model of a general financial crisis is presented in this chapter, while the various phases of the speculative manias that lead to crises are illustrated in the following chapters. This model of a crisis covers the boom and the subsequent bust and centers on the episodic nature of the mania and the subsequent events. This model differs from those that focus on the variations and the periodicity of economic expansions and contractions, including the Kitchin inventory cycle of thirty-nine months, the Juglar cycle of investment in plant and equipment that has a periodicity of seven or eight years and the Kuznets cycle of twenty years that highlights the rise and fall in housing construction.¹ In the first two-thirds of the nineteenth century, crises occurred regularly at ten-year intervals (1816, 1826, 1837, 1847, 1857, 1866), thereafter less regularly (1873, 1907, 1921, 1929).

The model

A model developed by Hyman Minsky helps explain the financial crises in the United States, Britain and other market economies. Minsky highlighted that the changes in the supply of credit were pro-cyclical and increased when the economy was booming and decreased during slowdowns. During the expansions investors became more optimistic and they revised upward their estimates of the profitability of a wide range of investments and became more eager to borrow. At the same time, both the lenders’ assessments of the risks of individual investments and their risk averseness declined and they became more willing to make loans, including some for investments that previously had seemed too risky.

When the economy slowed, investors became less optimistic and more cautious. The lenders also became more cautious as their loan losses increased, especially if the losses led to declines in their capital.

Minsky believed that the increases in the supply of credit in good economic times and the subsequent decline in the supply led to fragility in financial arrangements and increased the likelihood of a crisis. His model is in the tradition of the classical economists, including John Stuart Mill, Alfred Marshall, Knut Wicksell and Irving Fisher, who focused on the variability in the supply of credit. Minsky followed Fisher and attached great importance to the behavior of heavily indebted borrowers, particularly those that increased their indebtedness to buy real estate or stocks or commodities in search for short-term capital gains. Their motive was the profits from the increases in the prices of these assets, which they anticipated would greatly exceed the interest payments on the borrowed money. When the economy slowed many of these borrowers would become distress sellers because the prices of these assets would be falling.

Minsky suggested that the events that lead to a crisis start with a ‘displacement’ or innovation, some exogenous shock to the macroeconomic system.² If the shock was sufficiently large and pervasive, the economic outlook and the anticipated profit opportunities would improve in at least one important sector of the economy. Business firms and individuals would borrow to take advantage of the increase in the anticipated profits in this sector. Economic growth would quicken and in turn there might be a feedback to even greater optimism. It’s ‘Japan as Number One’ or the ‘East Asian Miracle’ or ‘The New American Economy’ or ‘the Geyser Tiger’ – much more profound optimism about the economy. The words differ across the countries but the tune is the same.

The shock varies from one speculative boom to another. The shock in the United States in the 1920s was the rapid expansion of automobile production and development of highways together with the electrification of much of the country and the large increase in households with telephones. The shocks in Japan in the 1980s were rapid increases in the supplies of money and of credit and financial liberalization that enabled the banks to increase their real estate loans at a rapid rate. The shock in the Nordic countries in the 1980s was financial liberalization, which permitted the domestic banks to borrow in the offshore market. One of the shocks the preceded the Asian financial crisis was the discovery of ‘emerging market equities as a new asset class’ which led to sharp increases in the purchases of these securities by mutual funds and pension funds headquartered in the United States, Britain, and other industrial countries. The shock in the United States in the 1990s was the revolution in information technology and the sharp declines in the costs of communication. The shock in the US housing market in the 2002 was securitization which involved the packaging of mortgages with similar attributes into bundles that provided the basis for issuing collateralized mortgage obligations (CMOs) and collateralized debt obligations (CDOs); the amount of money available for purchases of homes increased sharply. The shock in Iceland after 2000 was the privatization of the banks. At times the shock has been outbreak of war or the end of a war, a bumper harvest or a crop failure, the widespread adoption of an invention with pervasive effects – canals, railroads. Unanticipated changes of monetary policy like the one in the autumn of 1979 have also constituted major shocks.

If the shock is sufficiently large and pervasive, the anticipated profit opportunities improve in at least one important sector of the economy: the profit share of GDP increases. In the early 1980s, US corporate profits were 3 percent of GDP; toward the end of the 1990s this ratio had increased to 10 percent. The significant increase in stock prices associated with the dot.com boom at the end of the 1990s resulted in part from the much more rapid increase in corporate profits than in GDP.

The boom in the Minsky model is fueled by the expansion of credit. In the pre-banking seventeenth and eighteenth centuries personal credit or vendor financing fueled speculative booms. Once banks had been developed they expanded the supply of credit; in the first several decades of the nineteenth century banks increased the supplies of credit by issuing more notes and subsequently they increased the supply by adding to the deposit balances of borrowers. In addition new banks were formed and their efforts to increase market share led to rapid growth of credit because the established banks usually were reluctant to accept a decline in market share. In the 1970s the European banks began to poach on the turf of the US banks in making loans to the governments in Latin America.

One central policy issue centers on the control of credit from banks and from other suppliers of credit. Often the authorities in a country have applied strict controls that limited the ability of banks to make certain types of loans. The banks then set up subsidiaries that could make these loans, or their holding companies made these loans. Even if the credits from the financial institutions were controlled, increases in the supply from non-bank sources could lead to a boom.

Assume an increase in the demand for goods and services. Eventually the increase in demand presses against the productive capacity, prices increase and the more rapid increase in profits attracts both more investment and more firms. Positive feedback develops as the increase in investment leads to increases in the growth rate that in turn induces additional investment.

Minsky noted that ‘euphoria’ might develop at this stage. Investors buy goods and securities to profit from the capital gains associated with the anticipated increases in their prices. The authorities recognize that something exceptional is happening and while they are mindful of earlier manias, ‘this time it’s different’, and they have extensive explanations for the difference. The Chairman of the US Federal Reserve, Alan Greenspan, discovered a surge in US productivity in 1997, about a year after he first became concerned that US stock prices might be too high; the increase in productivity meant that profits would increase at a more rapid rate, and the higher level of stock prices relative to corporate earnings might not seem unreasonable.

The continuation of the process leads to what Adam Smith and his contemporaries called ‘overtrading’. This term is not precise and includes speculation about increases in the prices of assets or commodities, an overestimate of prospective returns, and ‘excessive leverage’.³ Speculation involves buying assets for resale at higher prices rather than for their investment income. The euphoria leads to an increase in the optimism about economic growth and about the increases in corporate profits. In the late 1990s Wall Street security analysts projected that US corporate profits would increase by 15 percent a year for five years. (If their forecasts had been correct, then the ratio of US corporate profits to US GDP would have been 40 percent higher than ever before at the end of the fifth year.) Loan losses incurred by the lenders decline because asset prices are increasing and they become more optimistic and reduce the minimum down payments and the minimum margin requirements. Even though bank loans are increasing, the leverage – the ratio of debt to capital or to equity – of many of their borrowers may decline because the rapid increase in the prices of real estate or securities leads to increases in their net worth.

A follow-the-leader process develops as firms and households see that the speculators are making a lot of money. ‘There is nothing as disturbing to one’s well-being and judgment as to see a friend get rich.’⁴ Unless it is to see a non-friend get rich. Similarly banks may increase their loans because they are reluctant to lose market share to other lenders. More and more firms and households begin to participate in the scramble for high profits. Making money never seemed easier. Speculation for capital gains leads from normal, rational behavior to what has been described as a ‘mania’ or a ‘bubble’.

The word ‘mania’ emphasizes irrationality; ‘bubble’ foreshadows that some values will eventually decline and probably sharply. The term bubble means any significant increase in the price of an asset or a security or a commodity that cannot be explained by the ‘fundamentals’. Small price variations around the values based on the fundamentals are ‘noise’. In this book, a bubble is an increase in indebtedness at a rate of 20 or 30 percent a year for an extended period that cannot be sustained. Someone with ‘perfect foresight’ should have foreseen that the price increases were not sustainable and that an implosion was inevitable.

In the twentieth century most of the manias and bubbles have centered on real estate and stocks. There was a real estate mania in southeast Florida in the mid-1920s and an unprecedented bubble in US stocks in the last three years of the 1920s; US real estate prices also increased significantly in late 1920s although not as rapidly as stock prices. In Japan in the 1980s the speculative purchases of real estate induced a boom in the stock market. Similarly the bubbles in the Asian countries in the early 1990s involved both real estate and stocks, and generally increases in real estate prices pulled up stock prices. The US bubble in the late 1990s centered on stocks, although the increases in household wealth in Silicon Valley and several other regions led to surges in real estate prices. The oil price shocks of the 1970s led to sharp increases in real estate activity in Texas, Oklahoma, Louisiana, and other oil-producing areas. Similarly the rapid increases in the prices of cereals in the inflationary 1970s led to surges in land prices in Iowa, Nebraska and Kansas and other Midwest farm states.

International propagation

Minsky focused on the pro-cyclical changes in the supply of credit in a single country. Historically euphoria has often spread from one country to others through one of several channels. The bubble in Japan in the 1980s had significant impacts on South Korea and Taiwan, which were parts of the Japanese supply chain; if Japan had been doing well economically, its former colonies also did well. Moreover the surge in asset prices in Japan had a dramatic impact on real estate prices in Hawaii, which is to Tokyo as Miami is to New York; Japanese travel to Hawaii for rest and recreation in the sun and the state experienced a real estate boom as the Japanese bought second homes and golf courses and hotels.

Arbitrage is one conduit – investors ensure that the increases in the price of a commodity in one national market lead to comparable increases in the prices of the same commodity in other national markets. Thus changes in the price of gold in Zurich, Beirut, and Hong Kong are closely tied to changes in the price of gold in London. Similarly increases in the prices of securities in one national market lead to nearly identical changes in the prices of the same securities in other countries. The prices of securities which are more distant substitutes for each other also are affected – the cliché that ‘the rising tide lifts all ships’ applies.

In addition increases in GDP in a large country induce increases in imports and hence increases in counterpart exports in other countries and in their GDPs. Capital flows constitute a third link; the increase in foreign purchases of securities available in a country lead to increases in both their prices and in the value of the country’s currency.

Moreover there are psychological connections, an increase in investor euphoria or pessimism in one country affects investors in others. The declines in stock prices on 19 October 1987 were practically instantaneous in all national financial centers (except Tokyo), far faster than can be accounted for by arbitrage, income changes, capital flows, or money movements.

In the textbook model an increase in the flow of gold and hence in the money supply in one country would be matched by a corresponding decline in the gold supply in a foreign country, and the increase in the money supply and in credit in the first country would be offset by the contraction of credit and money in the other. In the real world, however, the increase in the credit expansion in the first country may not be followed by a contraction of credit elsewhere, because investors in the second country may respond to rising prices and profits abroad by demanding more credit so they can buy the assets and securities whose prices they anticipate will increase. The potential contraction from the shrinkage in the monetary base in the second country may be overwhelmed by the increase in the speculative demand for credit.

As the boom continues, interest rates, the speed of payments and the commodity price level increase. The purchases of securities or real estate by ‘outsiders’ means that the ‘insiders’ – those who own these assets – sell them and realize profits – if the outsiders are buyers, then the insiders must be sellers. At every moment the purchases of real estate or stocks by the outsiders are necessarily balanced by sales of the insiders – usually at ever-higher prices. In 1928 the market value of the transactions on the New York Stock Exchange increased at an annual rate of 36 percent, and in the first eight months of 1929 at an annual rate of 53 percent. Similarly in 1998 the market value of the stocks traded on the NASDAQ increased by 41 percent and in the subsequent fifteen months the market values doubled. Investors rush to get on the train even as it accelerates. As long as the outsiders are more eager to buy than the insiders are to sell the prices of the assets or securities increase.

As the buyers become less eager and the sellers become more eager, an uneasy period of ‘financial distress’ follows; the corporate finance term reflects that a firm is unable to fulfill its debt servicing commitments. For the economy as a whole, the equivalent is the awareness by many firms and individual investors that it is time to become more liquid – to increase their money holdings. They sell real estate and stocks – their prices may fall sharply. Some highly leveraged investors may go bankrupt because the decline in prices is so large that the market value of these assets declines below their indebtedness – they have ‘upside down mortgages’. Some investors continue to hold the assets because they believe that the price decline is temporary, a hiccup.

The prices of the securities may increase again, at least for a while. There were four ‘bear market rallies’ in Tokyo in the 1990s; each involved increases in stock prices of more than 20 percent despite the downward price trend. But some investors believed that stock prices had declined too far, and they wanted to be among the first to buy the stocks while they were still ‘cheap’.

As the decline in prices continues, more and more investors realize that prices are unlikely to increase and that they should sell before prices decline further; in some cases this realization occurs gradually and in others suddenly. The race into money may turn into a stampede.

The specific signal that precipitates the crisis may be the failure of a bank or of a firm, the revelation of a swindle or defalcation by an investor who sought to escape distress by dishonest means, or a sharp fall in the price of a security or a commodity. The rush is on – prices decline and bankruptcies increase. Liquidation sometimes is orderly but may escalate into a panic as investors realize that only a relatively few can sell while prices remain not far below their peak values. The term ‘revulsion’ was used to describe this behavior in the nineteenth century. The banks become much more cautious in making new loans based on the collateral of commodities and securities and real estate because their values are more uncertain. In the early nineteenth century this condition was known as ‘discredit’.

‘Overtrading’, ‘revulsion’, and ‘discredit’ have a musty, old-fashioned flavor; they convey a graphic picture of the decline in investor optimism. Revulsion and discredit may lead to panic (or as the Germans put it, Torschlusspanik, ‘door-shut-panic’) as investors crowd to get through the door before it slams shut. The panic feeds on itself until prices have become so low that investors are tempted to buy the less liquid assets, or until trade in the assets is stopped because limits have been placed on maximum daily price declines, or because the exchanges have been closed, or because a lender of last resort succeeds in convincing investors that money will be available to meet the demand for cash. Confidence may be restored even without a large increase in the volume of money once investors conclude that credit is now more readily available.

Whether a lender of last resort should provide liquidity to forestall a panic and the decline in prices of real estate and stocks has been debated for more than two hundred years. Those who oppose the provision of liquidity from a lender of last resort argue that the knowledge that such credits will be available encourages speculation. Those who want a lender of last resort worry more about coping with the current crisis and reducing the likelihood that a liquidity crisis will cascade into a solvency crisis and trigger a severe recession – like the one that began in 2008. At the international level, there is neither a world government nor any world bank adequately equipped to serve as a lender of last resort. The International Monetary Fund has not met the expectations of its founders as a lender of last resort.

The validity of the model

Three types of criticism have been directed at the Minsky model. One is that each crisis is unique so that a general model is outdated. A second is that this type of model is no longer relevant because of changes in the economic environment. A third is that asset price bubbles are highly improbable because ‘all the information is in the price’ – the mantra of the efficient market view of finance.

Each criticism merits a response. The first criticism is that each crisis is the product of a unique set of circumstances, or that there are such wide differences among economic crises so they should be arranged into various species, each with its own particular features. Financial crises were frequent in the first two-thirds of the nineteenth century and in the last third of the twentieth century. In this view, each unique crisis is a product of a specific series of historical accidents – which was said about 1848 and about 1929,⁵ and may be inferred from the historical accounts of separate crises noted throughout this book. Each crisis also has its unique individual features – the nature of the shock, the object of speculation, the form of credit expansion, the ingenuity of the swindlers and the nature of the incident that touches off revulsion. But if one may borrow a French phrase, the more something changes, the more it remains the same. Details proliferate; structure abides. Most of these crises followed from the implosion of a credit bubble.

More compelling is the suggestion that the genus ‘crises’ should be divided into commercial, industrial, monetary, banking, fiscal, and financial (in the sense of financial markets) sub-species or into local, regional, national, and international groups. Taxonomies along such lines abound. This view is not accepted because the primary concern is with international financial crises that involve a number of critical elements – speculation, the increase in credit, an increase in the prices of securities or of real estate or commodities followed by a sharp fall as investors rush to sell. The test is whether the Minsky model provides insights about the broad features of crises.

The second criticism is that the Minsky model of the instability of the supply of credit is no longer relevant because of structural changes in the institutional features of the economy, including the rise of large corporations and of big labor unions and big governments, modern banking and speedier communications. The financial debacles in Mexico, Brazil, Argentina, and more than ten other developing countries in the early 1980s are consistent with the Minsky model; the external indebtedness of these countries increased much more rapidly than the interest rates on their loans. The bubble in real estate prices and stock prices in Japan in the second half of the 1980s and the subsequent implosion of asset prices is consistent with the Minsky model since the annual increases in the prices of stocks and of real estate were three or four times higher than the interest rates on the money borrowed to buy these assets. The booms and the subsequent busts in the second half of the 1990s in Thailand and Hong Kong and Indonesia and then in Russia feature the same pattern of cash flows.

The third criticism is that bubbles are impossible because market prices always reflect the economic fundamentals, and that sharp declines in asset prices usually reflect ‘policy switches’ by government or central banks. Those who take this position suggest that the alleged bubble appears to be the result of herd behavior, positive feedback or bandwagon effects – credulous suckers following smart insiders. These critics suggest that the model is ‘mis-specified’, that is, that something was going on and not taken into account by the theory, and that more research is needed.⁶ Some of the research ignored by those with this belief is offered in this book.

A more cogent attack on the Minsky model was by Alvin Hansen who claimed that the model was relevant prior to the middle of the nineteenth century but ceased to be so because of changes in the institutional environment.

Theories based on uncertainty of the market, on speculation in commodities, on ‘overtrading,’ on the excesses of bank credit, on the psychology of traders and merchants, did indeed reasonably fit the early ‘mercantile’ or commercial phase of modern capitalism. But as the nineteenth century wore on, captains of industry ... became the main outlets for funds seeking a profitable return through savings and investments.⁷

Hansen – who was a foremost expositor of the Keynesian model of the business cycle and especially of persistent high levels of unemployment – wanted to downplay the significance of alternative explanations for declines in economic activity other than a high level of saving. Hansen’s emphasis on the importance of the relation between savings and investment does not require the rejection of the view that changes in the supply of credit can have important impacts on the prices of securities and economic activity.

The model’s relevance today

The Minsky model can be readily applied to the currency market and to periods of overvaluation and undervaluation of the German mark, the Swiss franc, the Japanese yen, the euro, and other national currencies that are associated with ‘overshooting’ and ‘undershooting’. Changes in the values of national currencies have been large relative to long-run equilibrium values despite extensive intervention by central banks to dampen these deviations. Speculation in currencies has resulted in substantial profits for the large international banks although a few banks and industrial firms have incurred large losses.⁸

Consider the growth in the external debt of Mexico, Brazil, Argentina, and the other developing countries in the 1970s; bank loans to these countries increased by 30 percent a year and the external debt of these countries increased by 20 percent a year. The bank loans generally had a maturity of eight years and interest rates were floating and set with a specified markup over London Interbank Offer Rate (LIBOR). The interest rates averaged 8 percent although they trended upward during the decade. Since the cash that the borrowers received from new loans was substantially larger than the interest payments on their outstanding loans, they incurred no burden or hardship in making their debt service payments on a timely basis.

The inflow of foreign funds led to a real appreciation of the currencies of the money-receiving countries which was necessary so that the increase in their trade and current account deficits would match the increase in their capital account surpluses. It was inevitable and hence predictable that at some future date the receipts of money from new loans would decline below the interest payments on the outstanding loans, and at that time the value of their currencies would decline; the counterpart of the decline in the money inflows was that these countries would need trade surpluses to get some of the money necessary to pay the interest to their foreign creditors. Most of these borrowers effectively defaulted on their loans when the lenders stopped making new loans. These defaults cost the lenders $250 billion in the form of the reduction in the face value of the loans and what in effect was a reduction in the interest rates to below market levels. The lenders had failed to ask, ‘Where will the borrowers get the cash to pay us the interest if we stop supplying them with the cash in the form of new loans?’.

During the 1980s real estate prices and stock prices in Japan surged; the country experienced an economic boom. Real estate investors earned 30 percent a year – and much more after adjusting for leverage. Business firms recognized that the profits on real estate investments were substantially higher than the profits from making steel or automobiles or TV sets and so they bought real estate. Real estate prices were increasing many times more rapidly than rents. At some stage, the net rental income declined below the interest payments on the money borrowed to buy the real estate and the borrowers had a ‘negative carry’. The borrowers might obtain the funds to make the interest payments by increasing their loans against some of the properties that they already owned. At the beginning of 1990, the incoming governor of the Bank of Japan instructed the banks to limit the growth in real estate loans. Once the bank loans for real estate began to increase at a much less rapid rate, some of the firms and investors that needed money from new loans to pay the interest on the outstanding loans were no longer able to obtain new loans. They sold real estate and the bubble imploded.

The US international financial position after 2000 in some ways parallels that of Mexico, Brazil, and Argentina in the 1970s. These countries had unsustainably large current account deficits and obtained the money to pay the interest to their foreign creditors from new loans. The implication is that the US external payments position is not sustainable.

This book is a study in financial history, not economic forecasting. But investors seem not to have learned from experience.