The Collapse of Northern Rock

In 2007, Northern Rock was one of the five largest mortgage lenders in the UK. It had begun life as a building society, a peculiarly British cooperative institution that ploughed back all profits to its members. Traditionally, banks and building societies in the UK borrowed money from local savers. They took this money and lent it to local borrowers in the form of mortgages that were secured by residential property. The bank manager knew the customers and had a personal relationship with all of his clients. The building societies were owned by the savers, and any profits they made through spreads on lending and borrowing rates were returned to savers as dividends.

In the 1990s, Northern Rock was allowed by the government to convert itself into a profit-making institution and to sell shares on the stock exchange. In the early years of the new millennium, Northern Rock and other commercial banks began to make riskier loans and to borrow from each other on a short-term basis to provide the capital for their mortgages. Northern Rock began to provide mortgages worth 125% of the value of homes. Since it had a relatively small amount of deposits from savers, it relied instead on the ability to borrow money cheaply on the London Interbank Market to finance its loans.

The rate at which banks borrow and lend to each other is called LIBOR, the London Interbank Offered Rate. In August 2007, the LIBOR began to climb steeply and Northern Rock’s business model became unsustainable. It was forced to ask the Bank of England for emergency funds, and in February 2008, Northern Rock became the first of many world financial institutions to be owned, wholly or in part, by the taxpayer. Shortly following the fall of Northern Rock, the global financial system underwent a meltdown that hadn’t been seen since the 1930s. This book is about how we got to that point and what we can do in the future to prevent it from happening again.

Classical and Keynesian Economics

There is a major disagreement between two groups of economists about how the economy works. On one side, there are classical economists such as Eugene Fama of the University of Chicago, who believe that unregulated markets are inherently self-stabilizing and that government intervention often does more harm than good. On the other side, there are Keynesian economists such as the Nobel Laureate and New York Times columnist Paul Krugman, who believe that the market system needs a little help sometimes.

The Great Depression

The Great Depression caused a change in the political sphere that persists to this day. Western democracies began to recognize a vastly increased role for the federal government in the management of economic affairs, and following the Employment Act of 1946, U.S. politicians were given a much larger role in the management of the economy than they had previously enjoyed.

Why was the increased role for government accepted by the people? A major reason is that John Maynard Keynes provided a theoretical explanation of what had gone wrong. In his 1936 book, he explained what caused the Great Depression and he provided a remedy to prevent events like it from occurring again. The main difference of Keynes’s ideas from those of his predecessors was his rejection of the idea that the economy is a self-regulating system. The classical economists thought that the economy, if left to itself, would quickly return to full employment. Keynes disagreed.

Stagflation

Keynesian economics was widely accepted after World War II as a correct description of the way the economy works. From Keynesian theory there came a prescription for how to run policy that was followed successfully for three decades, from 1940 through 1970. In recessions, the central bank should lower the interest rate to stimulate private spending and increase aggregate demand. This is called monetary policy. In recessions, the government should spend more and pay for it through increased borrowing. This is called fiscal policy.

The loss of faith in Keynesian economics occurred as a consequence of a confluence of events in the 1970s that was unexpected because it was inconsistent with the basic tenets laid out by Keynes in The General Theory. In 1975, unemployment rose above 9% for the first time since the Depression and at the same time inflation rose above 13%. This coincidence of inflation and unemployment was dubbed stagflation by contemporary writers. Since Keynesian economics claimed that high unemployment and high inflation could not occur together, academic economists abandoned Keynesian theory. But although academic economists gave up on Keynes, economic policymakers did not give up on Keynesian policies.

Recently, politicians and commentators have rediscovered Keynes, and governments around the world have begun to spend money freely that they don’t have. The remarkable new move toward fiscal profligacy has led to discomfort among some academic economists who believe that the stagflation of the 1970s discredited the Keynesian theory that supports deficit spending as a way out of a depression. Important critics of fiscal deficits include Robert Barro of Harvard University and John Taylor of Stanford University. Barro and Taylor are classical economists who believe that government intervention often does more harm than good.

Other academic economists do not have fully workedout theories of the crisis but are willing to back the fiscal stimulus because they believe that Keynesian economics is sound and that there is no good alternative to save the global economy.

Why Fiscal Policy is the Wrong Approach

Keynes advocated fiscal policy because he thought that private firms were not investing enough during the Great Depression. He thought that consumption would go up automatically when income went up because people spend a fixed fraction of their income and save the rest. But two decades of research in the 1950s and 1960s showed that consumption does not depend on income: It depends on wealth. When government spends more, households save more. They know that the government will not be able to provide for their retirements in the future if it has a huge debt to repay. That is exactly what happened in the United States, the UK, and Europe in 2009 in response to the fiscal stimulus; the increase in saving partially offset the positive effect of the increased expenditure by government. Fiscal policy can help the economy out of the recession; but it is not nearly as effective as the Keynesians think, and the cost will be a permanent increase in the size of the government sector that will be paid for by our grandchildren.

Classical economics can be split into two parts: general equilibrium theory and the quantity theory of money. General equilibrium theory was developed in 1874 by a French economist, Léon Walras, who taught at the University of Lausanne in Switzerland. It explains how much of every good is produced and how the price of each good is set relative to every other good. For example, general equilibrium theory aims to tell us what determines how many hours will be worked by every person in the world, the number of cars produced in Japan, and the number of hours you would need to work to be able to afford a golfing holiday in Scotland.

The quantity theory of money was developed by David Hume, a Scottish philosopher and economist who was a leading figure in the Scottish Enlightenment and a contemporary of Adam Smith. Quantity theory is about money prices as opposed to real quantities and relative prices. It aims to tell us what determines how many dollars, pounds, or yen the average person will earn for an hour’s work, or the dollar, pound, or yen price you will have to pay for a car and the money cost of your hotel bill when you arrive in St. Andrews and tee up for the first hole. The quantity theory of money is also used to understand what determines the rate of inflation.

General equilibrium theory is built on the theory of demand and supply that was described in its modern form by Alfred Marshall in his book Principles of Economics. Marshall’s theory of demand and supply explained how much of any given commodity is produced and the price at which it is bought and sold. Walras applied the theory of demand and supply to all of the commodities in the economy at the same time. He asked the question: Is there a system of prices, one for every good, such that the quantities demanded and supplied of every commodity are equal simultaneously?

To answer Walras’s question, you need to account for all of the possible connections among markets. If the price of oil goes up, that will affect the demand for public transport and it will increase the number of taxis needed in London. Will there be enough workers to produce the extra taxis? At what wage? Under quite general conditions, economists have shown that there is at least one system of quantities and prices, including wages and labor allocated to each industry, under which the demands and supplies of all commodities are in balance all at once. Walras showed that capitalism can work in theory. But does it work in practice?

Walras’s successor at the University of Lausanne was Vilfredo Pareto. Pareto was not just concerned with the existence of an economic equilibrium. He wanted to know if free markets allocate resources among different members of society in the best possible way.

Pareto is known to economists for asking the question: Could an omniscient planner reallocate commodities among all the different people in the world in a better way than the free market? Under Pareto’s definition of “better,” the answer was a resounding no. In other words, markets do just about as good a job of allocating resources as anyone could possibly imagine. Although Pareto’s work implies that free markets work well, this proposition rests on a set of assumptions about the properties of technology and the way people behave. By laying out these assumptions, Pareto provided the groundwork for later economists to ask a related question: Under what circumstances does the free market break down? Pareto’s concept of a better allocation of commodities is so important that it has been formalized in economics in the form of a theorem called the first theorem of welfare economics. The first welfare theorem says that there is a connection between Pareto’s notion of a better allocation of commodities and the way that commodities are allocated in market economies. Pareto’s notion of a better allocation says nothing about who is rich and who is poor; he takes this as given. It simply says that markets allocate resources efficiently given the existing distribution of wealth. Economists call the notion of a better allocation, embodied in Pareto’s work, Pareto efficiency. The first welfare theorem is the statement that every competitive equilibrium is Pareto efficient.

David Hume and his contemporaries developed the quantity theory of money, which asserts that the money value of all of the goods and services produced in a given year (the money value of gross domestic product, or GDP) is proportional to the stock of money. The classical economists used general equilibrium theory to determine the physical quantities of goods and services that would be produced and to determine their prices relative to each other. These are called relative prices. They used the quantity theory of money to determine the average level of prices in terms of money. These are called money prices or nominal prices.

The difference between relative prices, determined by general equilibrium theory, and money prices, determined by the quantity theory of money, is this. Suppose that Denis, the last worker hired by the Aluminum Box Company, can produce 40 boxes per week by working for 40 hours. Classical general equilibrium theory predicts that the wage paid to Denis would be one commodity per hour. It has nothing to say about the money wage or the money price of an aluminum box. These money prices remain unexplained by the theory. In equilibrium, it might be that Denis earns $1 an hour and an aluminum box costs $1, or Denis earns $10 an hour and an aluminum box costs $10. Both situations are equally consistent with the theory of relative prices contained in general equilibrium theory. That’s where the quantity theory of money comes in. According to the quantity theory, the dollar price of a commodity depends on how many dollars are circulating since dollars are needed to facilitate exchange. The idea that money is the oil that keeps the wheels turning is very powerful, and some version of it is built into every modern interpretation of classical economics. At its core is the notion that real economic activity is determined by the fundamentals of the economy—preferences, endowments, and technology—and that, in the long run, the quantity of money determines only the price level. This idea is expressed by the proposition that, in the long run, money is neutral.

Helicopter Ben

Friedman asked us to consider what would happen if a helicopter were to fly over a country and drop dollar bills. A crude rendition of the quantity theory would assert that the effect would be an immediate doubling of all money prices and all quantities traded would remain unchanged. Everybody would wake up on Wednesday morning, after a helicopter drop on Tuesday night, and a cup of coffee that cost $1 on Tuesday would cost $2 on Wednesday. But the theory is much more sophisticated than that. Although it asserts that money has no long-run effect on employment or output, the quantity theorists do not make that assertion about the short run because the evidence suggests otherwise. In practice, it takes time for the effect of an increase in the stock of money to work its way through the economy.

There is a tension between general equilibrium theory and the quantity theory of money that persists in economics to this day. General equilibrium theory predicts that relative prices will be determined to equate the quantities demanded and supplied for all commodities simultaneously. The quantity theory of money predicts that the general level of prices will be proportional to the stock of money. But when new money enters the economy, all prices do not change overnight. Hence, during the transition from the short run to the long run, the predictions of the general equilibrium model and the long-run predictions of the quantity theory cannot both be true. Economists attribute the difference between the two theories to the effects of economic frictions.

Economics Frictions

Economists use the word “friction” to mean a restriction on trade or a cost of changing a price that prevents firms from adjusting wages and prices quickly to the levels that are predicted by general equilibrium theory. Economists invoke economic friction to explain why all prices do not change overnight in response to an increase in the quantity of money.

When the model of demand and supply is applied to the labor market, it implies that the quantity of labor demanded should be equal to the quantity of labor supplied. Because we often observe high unemployment, economists argue that this unemployment must be due to an economic friction that slows down the adjustment of the wage. Large frictions imply that prices and wages adjust slowly and so there is a higher chance of seeing high and persistent unemployment.

MayNard Keynes’s New Vision

Keynesian economics was different from everything that went before. In the 1920s, the classical economists saw the economy as a stable self-correcting machine. Random unpredictable events might cause a disturbance to the economy that would temporarily throw some people out of work. But hundreds of millions of selfish individuals would be guided, in the words of Adam Smith, “by an invisible hand,” to move the economy quickly back to full employment.

Keynes was much more skeptical of the self-correcting nature of the economy because he saw no evidence of it in Great Britain during the 1920s, when unemployment remained high for a decade. In his view, one of the most important shocks to the economy is a shock to the confidence of investors about the future value of the stock market. He called this the “animal spirits of investors.” When combined with Keynes’s theory of the labor market, the possibility that markets may be driven by confidence implied that very high unemployment could persist for a very long time for no good reason.

Unemployment During the Great Depression

Business cycles are not new. The United States underwent five major financial crises in the nineteenth century, and all of them had some elements in common. British economist Arthur Pigou summarized contemporary views of the theory of business cycles in an influential book, Industrial Fluctuations, published in 1929. He listed at least six different causes of business cycles including errors of optimism and pessimism, agricultural fluctuations caused by the weather, shocks to productivity as a consequence of new inventions, monetary fluctuations, industrial disputes, and changes in tastes. Nobody at this time disputed the fact that, left to itself, the economy would quickly return to full employment after one of these six factors caused a disturbance. The Great Depression of the 1930s changed this view forever.

Figure 3.2 plots the percentage of unemployed persons in the United States from 1890 through 2007. The ups and downs that occur at irregular intervals are a manifestation of business cycles, and it is these ups and downs that Pigou attributed to a laundry list of possible causes, from optimism and pessimism to changes in tastes. There are two features of the graph worth noting. First, the upward spike that began in 1929 and ended in 1941 is much larger than any spike in unemployment that has occurred before or since. The closest episode is the recession that occurred in the last decade of the nineteenth century, in which unemployment reached 18% and exceeded 10% for six years in a row. Second, fluctuations in the unemployment rate since 1946 have been less volatile than fluctuations before World War II. Keynesian economists claim, and I think they are right, that the reduction in the size of fluctuations in the United States after World War II is a direct consequence of the increase in the role of government that followed when Congress passed a new piece of legislation, the Employment Act of 1946, that encouraged the government to “promote maximum employment, production, and purchasing power.” The increased stability of postwar business cycles in the United States is direct evidence for the success of Keynesian economics.

Keynes’s Escape from Classical Economics

Classical ideas at the time were summarized by the theory laid out by Pigou in Industrial Fluctuations. According to these ideas, the Depression must have been caused by one of the six fundamental factors listed previously: These included errors of optimism and pessimism, agricultural fluctuations caused by the weather, shocks to productivity as a consequence of new inventions, monetary fluctuations, industrial disputes, or changes in tastes. Some modern business cycle theorists go further. They would claim that all shocks are caused by fundamental changes and most of these are due to the introduction of new technology: But whatever the origin of the shock, classical theory implies that the economy should quickly return to full employment.

According to Keynes, the impulse that caused the Great Depression was a spontaneous fall in confidence about the future—a kind of mass hysteria affecting all stock market participants simultaneously. It was the stock market crash that caused the Great Depression. How did this work? The stock market fell because people believed that the machines and factories that produce profit would, in the future, have a much lower value. Firms stopped buying new capital equipment and the workers who produced capital goods became unemployed. These workers stopped buying consumer goods and workers who produced these goods also lost their jobs. The economy did not return to full employment because there is no self-correcting mechanism of the kind envisaged by Pigou and his contemporaries. In Keynes’s view, any unemployment rate can persist forever because the forces that tend to restore equilibrium are either nonexistent or so weak that we would not expect to see them operating in finite time.

There are two major problems with this position. First, it is difficult to identify a fundamental shock of significant importance that could have triggered a depression of the magnitude that was experienced in the 1930s. Second, whatever this shock might have been, the millions of workers who lost their jobs in the early years should have quickly found new employment. The reality was very different and the cost in terms of human misery was incalculable.

Keynesian Theory

According to Keynes, the impulse that caused the Great Depression was a spontaneous fall in confidence about the future—a kind of mass hysteria affecting all stock market participants simultaneously. It was the stock market crash that caused the Great Depression. How did this work? The stock market fell because people believed that the machines and factories that produce profit would, in the future, have a much lower value. Firms stopped buying new capital equipment and the workers who produced capital goods became unemployed. These workers stopped buying consumer goods and workers who produced these goods also lost their jobs. The economy did not return to full employment because there is no self-correcting mechanism of the kind envisaged by Pigou and his contemporaries. In Keynes’s view, any unemployment rate can persist forever because the forces that tend to restore equilibrium are either nonexistent or so weak that we would not expect to see them operating in finite time.

Keynes’s theory of what went wrong in the Great Depression is based on the twin concepts of aggregate demand and aggregate supply. Aggregate demand and supply are similar to, but distinct from, Marshall’s theory of demand and supply in a single market. It is important not to confuse the two concepts. Aggregate demand explains the total money value of goods and services that all households and firms would like to spend in a given period of time. Aggregate supply explains how many workers are needed to produce the goods and services necessary to meet that demand.

Keynesian Policy

In contrast to this classical view that government should try to balance its budget in a recession, Keynes argued instead that the government should borrow money and use it to stimulate aggregate demand. He explained why this was appropriate with his new economic theory.

In classical economics, every dollar spent by government is one less dollar spent by households because the size of the pie is fixed. In Keynesian economics, an extra dollar spent by government increases the size of the pie and causes an increase in the amount available to both government and households. Keynes argued that, to escape from a depression, national governments should borrow and use the borrowed money to purchase goods and services from private firms. This theory of why governments should run fiscal deficits was used by President Nicolas Sarkozy in France, Chancellor of the Exchequer Alistair Darling in the UK, and Treasury Secretary Timothy Geithner in the United States to justify huge increases in public sector borrowing in these countries in 2008–2009.

Keynesian economics was discredited by the rise and fall of an idea: the Phillips curve. It was introduced to economics in 1958 by a New Zealander, Alban W. Phillips, known as Bill Phillips to his friends and family. Bill Phillips was an engineer by training. His enduring contribution to the literature of economics was to point out that there had been an inverse relationship between unemployment and the rate of change of money wages in a century of UK data. Historically, when unemployment had been low, wages would rise. When it had been high, they would fall. Keynesian economists embraced the Phillips curve, which they saw as empirical evidence in support of Keynesian economics. They argued that the government must choose high inflation or low unemployment by picking a point on the Phillips curve. But when the theory was put to the test, the predictions of the Keynesians failed dramatically. In place of the Phillips curve, economists introduced a new idea. The economy gravitates toward a natural rate of unemployment that cannot be influenced by fiscal or monetary policy.

Bill Phillips and His Curve

But Bill Phillips is best known for his work on inflation. He found that in the UK data, money wages fell when unemployment was high and they rose when it was low. He graphed this relationship between unemployment and wage inflation on a diagram that now bears his name—the Phillips curve. His work was important, because the same relationship that existed between inflation and unemployment in the 1860s also characterized the relationship between these two variables in the 1950s. The Phillips curve remained structurally stable for over a century. This was a revelation. If unemployment had always been high in the past, when inflation was low, economists argued that this same relationship should hold in the future. If government could influence one of these variables, it could also influence the other. The fact that the Phillips curve was structurally stable was thought to represent a feature of the private economy that represented a constraint on economic policy. Its existence was explained by the Keynesian theory of aggregate supply. If this theory is correct, increased fiscal deficits during the administrations of Presidents Kennedy, Johnson, and Nixon should have restored full employment without causing inflation.

Two American Keynesians

Two American Nobel Laureates, Paul Samuelson and Robert Solow, summarized the mainstream Keynesian view that evolved from Phillips’s article. They phrased the debate as a policy dilemma. If government attempted to control inflation by managing aggregate demand, the economy would face a trade-off. Lower inflation could be obtained at the cost of higher unemployment or lower unemployment at the cost of higher inflation. The government would have to choose which of the two alternatives was preferable. If government were to stimulate aggregate demand by lowering the interest rate or reducing taxes, it would lower the unemployment rate. But the cost of this policy would be higher inflation. Alternatively, if government were to reduce aggregate demand by raising taxes or increasing the interest rate, it would reduce inflation at the cost of higher unemployment.

Samuelson and Solow’s own work on inflation and unemployment in the United States confirmed that the same relationship held between inflation and unemployment in the U.S. data that Bill Phillips had discovered in the UK data. They interpreted the Phillips curve as a constraint on policy. For them, it was the job of the economist to find out how much inflation would be needed to reduce unemployment by a given amount. It was the job of the policymaker to choose how much unemployment and how much inflation society could tolerate. More of one would inevitably mean less of the other.

The Natural Rate Hypothesis

The Samuelson-Solow interpretation of the Phillips curve did not go unchallenged. In 1968, two influential Nobel Laureates, economists Edmund Phelps of Columbia University and Milton Friedman of the University of Chicago, argued in separate articles that we should not expect to see a permanent long-run relationship between inflation and unemployment. They explained this position by pointing out that the unemployment rate depends on fundamental real factors such as the productivity of workers, the preferences of households, and the time and trouble spent by workers in searching for jobs.

Friedman is saying that it is possible to augment general equilibrium theory by adding elements such as the cost of searching for a job, and that the augmented model would be able to explain why there is unemployment. A model that is augmented in this way would not display a long-run trade-off between unemployment and inflation because the factors that determine the unemployment rate have nothing to do with the quantity of money or the rate of inflation. Friedman’s argument asserts that Samuelson and Solow were wrong to view the Phillips curve as a trade-off that could be exploited by a policymaker to choose either high inflation and low unemployment or low unemployment and high inflation. Instead, he claimed that there is a natural rate of unemployment that is independent of fiscal and monetary policy and is consistent, in the long run, with any rate of inflation.

Phelps and Friedman did not have long to wait before events provided a dramatic confirmation of their thesis. During the 1970s, the U.S. economy experienced high inflation and high unemployment at the same time and the data did not lie anywhere near the Phillips curve that had characterized the relationship between these two variables over the previous century.

Bob Lucas and Economic Policy

Lucas was the leading figure in the rational expectations revolution, a pathbreaking new movement that swept macroeconomics in the 1970s and replaced Keynesian economics with an updated and revised version of classical ideas. Rational expectations economics uses sophisticated mathematics to provide a rigorous foundation to classical theory. It has had a profound effect on the discipline, and the theory of rational expectations dominates the curriculum at all major universities throughout the world today.

When Phelps and Friedman published their work on the natural rate hypothesis in 1968, most economists saw it as a statement about what would happen to the unemployment rate in the long run. They argued that if the Fed were to lower the interest rate, it might lower the unemployment rate over short periods of time of one to two years, but in the long run the unemployment rate would be determined by three fundamentals of the economy: the preferences of households, the stock of skilled and unskilled labor, and the current state of technology.

In 1972, Lucas published an important article in the Journal of Economic Theory. He took the arguments of Phelps and Friedman one step further by formalizing their ideas and adding a theory of how people form expectations. Rational expectations is the idea that you can’t fool all of the people all of the time. Whatever market participants believe about the future must be consistent, on average, with what happens. Lucas’s ideas led to the development of real business cycle theory, a movement that replaced Keynesian economics as the dominant approach used by macroeconomists to understand the real economy. It also led to new-Keynesian economics, an extension of real business cycle theory that adds economic frictions to understand the short-run and long-run effects of money on prices and employment. New-Keynesian economics explains why Fed policy influences unemployment first, and inflation much later.

Although Phelps and Friedman believed that the Fed could not alter the unemployment rate in the long run, they still believed that the Fed could improve economic welfare in the short run by lowering the interest rate in a recession. Lucas argued that policymakers cannot improve the welfare of the average citizen through monetary or fiscal policy even in the short run. His theory implied that unemployment can deviate from its natural rate only if households and firms make mistakes in their forecasts of future business conditions. The rational expectations revolution was a new way of thinking about macroeconomics that swept away Keynesian economics from most academic departments. It was successful because it came at a time when the public had lost faith in Keynesian ideas after economists failed to predict stagflation in the 1970s.

The French Influence

Lucas’s work is based on developments of general equilibrium theory by a French-born Nobel Laureate and naturalized American, Gérard Debreu, who spent most of his career in the economics department at Berkeley. He was trained as a mathematician at the École Normale Supérieure in Paris and he brought the rigor of mathematics to general equilibrium theory, as well as a huge amount of insight.

In a slim little volume, Theory of Value, Debreu took Walras’s concept of general equilibrium theory and used modern mathematics to provide very general conditions under which the first welfare theorem of economics is true.2 Almost as an afterthought, he added a chapter that argued that general equilibrium theory is a great deal more general than usually assumed. By thinking of a commodity as having a date of delivery and a geographical location, he showed that the theory could be used to understand all of trade at all points in time.

In an astonishing insight that has been hugely useful to the modern theory of finance, he went one step further. Commodities are different not just because they are delivered on a different date in a different location but also because of differences in random events that may or may not occur. Suppose that the weather in London could be sunny or rainy. An umbrella purchased in London on September 28, 2009, if the sun is shining is a different commodity from an umbrella purchased in the same place at the same time when it’s raining. Wall Street uses this idea to price securities by determining their values in terms of the underlying risk associated with any sequence of future payments.

Real Business Cycle Theory

The most influential figure in the development of real business cycle theory is the Nobel Laureate Edward Prescott. Together with fellow Nobel Laureate Finn Kydland, Prescott extended the methods pioneered by Lucas to describe macroeconomics.

Although real business cycle theory is mathematically rigorous, it is simpler than the verbal theories of classical business cycle theory described by Pigou. There is no unemployment in the model, and all variations in employment are voluntary variations in the number of hours that people want to work. This simplification allowed the developers of real business cycle theory to concentrate on what they believed to be its most important aspects: a description of the dynamics of employment, investment, consumption, and GDP that are triggered by changes in productivity that arise from the ebb and flow of new ideas as they impact the economy. As an example of the kind of shocks emphasized by real business cycle theory, consider the invention of the personal computer. This triggered a movement of workers out of old industries such as steel and automobiles and into high-tech industries that required different skills. During this process, employment fell temporarily but eventually, as new workers were trained, the economy became more productive and new jobs were created. The invention of the personal computer is an example of a shock that has effects that influence employment, consumption, and investment over a period of time, and the study of this process became the new standard for the study of booms and busts.

New-Keynesian Economics

The new-Keynesians chose that name to differentiate themselves from the new-classical economics of Lucas and Prescott. Lucas showed that monetary and fiscal policy cannot improve welfare, even in the short run. In real business cycle theory, it is costless for firms to change prices and wages. The new-Keynesians added an explicit cost of changing prices. By adding a friction of this kind, they hoped to overturn the new-classical result that policy cannot improve welfare and to show instead how government monetary and fiscal policy can improve people’s lives.

Since the time of Hume, economists have known that when money enters the economy, it first affects quantities and only later affects prices. This is widely understood by central bankers today. In 2009, the Fed responded to the recession by pumping money into the economy. This policy was based on new-Keynesian ideas. Initially, the Fed’s actions appear to have helped to prevent the recession from turning immediately into a full-fledged depression.

Greenspan (former chairman of the Fed) was worried about inflation because between September 2008 and March 2009, the Fed doubled its balance sheet by pumping $800 billion of new money into the U.S. economy. Historically, monetary expansion of this kind had led to inflation with a lag of more than three years. As of the summer of 2009, it was unclear whether inflation would reappear.

The way that money influences the economy, first by changing quantities and later by changing prices, is called the monetary transmission mechanism. The formalization of the monetary transmission mechanism using the ideas of the rational expectations revolution is a considerable intellectual achievement. But it doesn’t help us understand the Great Depression or the 2008 financial crisis.

Who Owns the Fed?

The Fed is managed by a seven-member board of governors, appointed by the president and confirmed by the U.S. Senate, and although the members of the board are political appointees, they serve for overlapping 14-year terms and once appointed cannot be removed from office. This gives the Fed a fair amount of autonomy from any given political administration, since individual appointees serve terms that may span several administrations. The primary role of the Fed is to manage the nation’s money supply.

Fighting Inflation

The Fed can choose to fight the recession by lowering the interest rate, but since recessions are temporary, this policy is one that generates only short-term gains. When the Fed lowers the interest rate to stimulate the economy in a recession, it may feed inflationary expectations. As the economy emerges from the recession, it does so at the cost of permanently higher inflation. This is what happened in the period from 1951 through 1982, as after every recession the inflation rate crept a little higher. Higher inflation went hand in hand with higher interest rates, and by 1981, the interest rate on a 30-year fixed-rate mortgage was over 18% per year. The effect on first-time home buyers was catastrophic. A GI returning from World War II would have paid $59 a month in interest and principal for every $10,000 dollars that he borrowed. By 1981, his children would be paying $150 for the same loan.

Fighting Unemployment

Figure 6.1 illustrates the history of the U.S. unemployment rate and the interest rate from 1951 through 2008. The unemployment rate is the line marked by circles and is measured on the left axis. The interest rate is the solid line measured on the right axis. The shaded regions are recessions defined by the Business Cycle Dating Committee of the National Bureau of Economic Research. Including the current recession, which began in December 2007, there have been 10 recessions since 1951. The figure illustrates clearly that every one of them was associated with a sharp increase in the unemployment rate and at the same time a sharp reduction in the interest rate on Treasury bills. In each recession, the interest rate fell because the Fed was trying to alleviate the adverse effects of the recession on people’s lives.

The Future of Central Banking

Milton Friedman asserted that long-run Fed policy should concentrate on price stability and that it can only reduce unemployment in the short run at the cost of increasing inflation in the long run. As long as the Fed restricts itself to the interest rate as its only policy instrument, Friedman is correct. But what if there were another demand-management tool that allowed the Fed to change the long-run unemployment rate? I believe that this tool exists—direct central bank intervention in the stock market to prevent bubbles and crashes by changing the composition of the central bank balance sheet. A second tool of this kind would provide a way of managing the long-run unemployment rate. This instrument is distinct from the primary tool of interest rate control that is used by central banks throughout the world to manage inflation. I will return to this idea in chapter 11, where I will discuss the practical aspects of implementing the management of aggregate stock market wealth.

Sand in the Oild

A job must be produced from inputs, just like any other commodity. Like a car that is produced from capital and labor, a job is produced from the time spent searching for each other by a firm and a worker. The physical description of this process is called the search technology. Although the production technology and the search technology are similar (see Figures 7.1 and 7.2), the inputs to the search technology are not like labor and capital. They are different because each of them is a commodity that one side of the market has more information about than the other.

Figure 7.1 illustrates the production process for a good. The inputs are capital and labor. Workers operate machines and add value to raw materials. The output is a good such as a car, a computer, or a can of beans. Figure 7.2 illustrates the production process for a job. The inputs are the search time of an unemployed worker and the search time of a recruiter in the personnel department of a firm. The recruiters sort through applications of unemployed workers and filter out those who are suitable for vacant jobs. Unemployed workers fill out applications and attend job interviews. The output is a match between an unemployed worker and a vacant position.

General equilibrium theory predicts that the search technology should be run by headhunting firms. Consider an example where the ABC employment agency matches workers with jobs. How would this work? The ABC employment agency should buy the rights to match unemployed workers and vacant jobs from households and firms and put together workers and firms in the same way a computer dating service matches lonely hearts. Once the ABC company has established that Maynard Jagger, an unemployed welder, would fit well with the vacant position at, the Detroit Auto Corporation, Detroit Auto Corporation should offer the job to Maynard and pay ABC for the information that he is the best fit.

Why Search Markets don’t Work Well

Why don’t private headhunting firms spring into existence to find jobs for unemployed workers? Although we do see some headhunting firms, they are a small fraction of the employment market, and they do not operate in the way that classical economic theory suggests they should. They operate as personnel departments for firms that are too small or too specialized to run their own operations. But they do not pay workers and firms for the right to match them because the markets for search inputs do not exist. It is not hard to see why.

How would these markets operate if they did exist? A dishonest unemployed worker could turn down every job he or she was offered and continue to receive payments while remaining unemployed. Since there will often be good reasons to refuse a job, it would be impossible to write a contract in which the worker must take any job that he or she is offered.

Why High Unemployemnt Exists?

A given number of jobs can be filled by a large number of unemployed workers and a few recruiters or by a few unemployed workers and a large number of recruiters. But should society match workers with jobs by asking a few unemployed workers to search for a lot of vacant positions or a lot of unemployed workers to search for a few vacant positions? Either outcome can occur in the real world because the price signals that should tell firms and workers how to behave are missing.

Economists say that any unemployment rate is an equilibrium, by which they mean that there are no forces acting to change the unemployment rate and that, as a consequence, very high unemployment can persist for a very long time.

Why The Wage Doesn’t Fall?

Keynesian economics is often criticized because Keynesians can’t explain why, when there are many unemployed people, the wage does not fall to restore equality between the quantity of labor demanded and the quantity supplied. Surely these unemployed people could offer to work for a lower wage than existing workers. It would be in the interest of a firm to hire an unemployed worker because it would be able to make a profit from the transaction. But this argument doesn’t take account of the costs of matching a worker with a vacant job.

Many unemployed workers looking for a job is like many fishermen in a common pond. When there are many other fishermen, it becomes harder for each one to catch a fish. Because no one owns the pond, there is no price signal to tell some of the fishermen to leave. The job search process is like this. There is no price signal to tell some firms to put more or fewer resources into recruiting, and the market sometimes gets the allocation between vacancies and unemployment very wrong. The behavior of other firms in the economy makes each individual firm more or less productive.

In the fall of 2008, the three major U.S. automakers faced tough times. Demand for automobiles had fallen dramatically; GM responded by cutting 15% of its salaried workforce, approximately 5,000 white-collar jobs in the last two months of 2008. Many of these salaried workers were from the personnel department. They were no longer needed because GM responded to decreased sales by producing fewer cars. The net response was a smaller, leaner auto industry with higher productivity and higher real wages. This is exactly what happened to manufacturing industries in the first three years of the Great Depression.

Do Fundamentals Drive Markets?

Classical economists argue that the value of the stock market is determined by fundamentals. According to this view, if Microsoft shares drop in value, it is because rational investors anticipate that Microsoft’s profits will fall. Perhaps there is a new competitor in the market. Perhaps there is a new invention that makes the personal computer obsolete. All market movements arise from rational investors anticipating changes in future fundamentals, and there is no room for the confidence of stock market participants to independently influence economic activity.

For nearly 70 years, investment advisors recommended the wise strategy of investing in the stock market for the long haul. Stocks had outperformed bonds by almost 5% on average over every 10-year period since records began. But in 2008, markets worldwide lost 40% of their value for no sound fundamental reason, and suddenly the premium on stocks made sense as 30 years of capital gains were wiped out overnight. Financial journalists declared that the 2008 crash was the death knell of the efficient markets hypothesis.

Behaviorl Economics or Rational Choice?

Keynes did not embrace rational expectations. But my goal is not to resuscitate failed interpretations of Keynes’s General Theory. It is to build on his key ideas. Keynes never showed how his theory fits with classical economics. His followers assert that unemployment persists because wages and prices are slow to adjust to clear markets. But the money wage fell by 30% between 1929 and 1932. Unemployment does not persist because wages are inflexible. Unemployment persists because it is costly to match workers with jobs and there are no price signals to tell firms the best way to do this.

Before we get even further into debt as a society, it is important to understand why fiscal policy failed in the 1970s in a way that is consistent with evidence from the Great Depression. Appealing to behaviorist assumptions to justify Keynesian economics does not provide us with the explanation we seek. We need a new theory that integrates classical and Keynesian ideas.

Wealth Matters

The Great Depression occurred because firms could not sell enough goods to maintain full employment. Unemployed workers didn’t have the purchasing power to buy goods, and that led to a vicious cycle that lasted for a decade. Keynes thought that aggregate demand depended mainly on income, but research on consumption in the 1950s proved that hypothesis to be incorrect. In fact, people spend more or less on goods and services based on wealth. Friedman (1957) showed that transitory fluctuations in income have a minor effect on consumption. What matters are sustained changes that he called permanent income, and other researchers identified with wealth. It follows that household wealth is a critical factor in determining the aggregate demand for goods and, in turn, the value of employment. This fact has important consequences for the effectiveness of a fiscal stimulus.

During the 1930s, many families owned tangible wealth in the form of bank accounts, but these assets disappeared as banks failed. Today, bank deposits are insured by the federal government, but direct ownership of risky assets in retirement accounts or direct ownership of stocks has become much more common. Wealth plays an important role in determining consumption, and when wealth falls and stays low for a protracted period of time, households are likely to cut back on their consumption, and aggregate demand for goods and services will fall.

Where Keynesian Economists Went Wrong?

Keynesian economists stress income as the main determinant of consumption. But although fluctuations in income are a factor in determining consumption, they are not the most important one. People recognize that fluctuations in income are often temporary. When income falls for six months because the breadwinner is between jobs, the household can often borrow against accumulated assets to maintain consumption. But when a person stays unemployed for a couple of years, his or her immediate sources of wealth quickly become exhausted.

It was a sustained drop in aggregate wealth that led to the depths and extent of unemployment at the time of the Great Depression, and it is a sustained drop in wealth that threatens to turn the 2008 crash into a very painful event. By stressing the role of income as a determinant of consumption, instead of wealth, Keynesian economists are led to advocate fiscal policy as the most effective remedy to restore full employment. I believe that they are wrong, and I am afraid that this mistake may be very costly since it will lead governments to accumulate large debts that our grandchildren will be asked to repay.

Where Classical Economists Went Wrong

Many classical economists still see the economy as basically sound, and it is not uncommon to hear the government blamed as the source of all of our current problems. When the Fed chairman, Ben Bernanke, and the Treasury secretary, Henry Paulson, called on the government to provide $800 billion in emergency funding to help bail out troubled financial institutions, many economists cried foul and they blamed the crisis on the government itself; a colleague of mine asserted in a private conversation that “If only Bernanke and Paulson had left the banks to fail, the system would quickly have reestablished an equilibrium at full employment. Instead, Bernanke cried fire in a crowded theater and the response was predictable.” Views similar to this have been voiced by many economists in recent days and are shared by a considerable number of academics, including some Nobel Prize winners.

I believe that the classical view is wrong for two reasons. First, modern classical economists ignore the role of confidence as an independent factor that drives booms and busts. Second, classical economists see the economy as a self-correcting mechanism in which market forces will restore full employment. They are wrong on both counts.

Two Black Mondays

The U.S. Great Depression began on Monday, October 28, 1929, when the stock market fell 13%, its second worst one-day decline in history. The worst was also on a Monday in October, but it did not occur for another 58 years. Both days have been described in the popular press as Black Monday.

To understand why many economists are worried that the 2008 Wall Street crash may have bad consequences for Main Street, take a look at Figure 9.1, which graphs the value of the S&P 500 and the unemployment rate for the early years of the Great Depression. The S&P 500 is the line marked by circles and is plotted on the left axis. The unemployment rate is the solid line plotted on the right axis. The unemployment rate is not available on a monthly basis for that period, which is why the unemployment series moves up in steps. Immediately following Black Monday in 1929, the U.S. unemployment rate began to climb from less than 8% in 1929 to 24% in 1932.

But although a big one-day decline in the stock market was followed by a depression in 1929, it is not true that big declines in the market are always followed by big declines in economic activity. On Monday, October 19,1987, the second and larger Black Monday in U.S. history, the S&P 500 dropped 21%, falling from 283 to 225 in one day, but this calamitous drop did not have much of an effect on employment. Why was 1987 different from 1929?

Figure 9.2 plots the S&P 500 and the unemployment rate for the period from 1986 through 1990. The S&P index is marked by circles and is measured on the left axis. The monthly unemployment rate is the solid line plotted on the right axis. This figure shows that although the market lost 21% of its value in one day, the decline did not last long and served only to wipe out gains that had built up over the previous year. Contrast this with the situation in 1929, where the S&P lost a third of its value in a month and continued to decline to 12% of its 1929 peak, a value that was lower in absolute terms than at any date since 1898.

GreenSpan the Wizard

The fact that the market recovered relatively quickly in 1987 was due in part to the actions of the Fed. Alan Greenspan was aware of the history of the events of the 1930s and he responded to the huge drop in the stock market by announcing publicly that the Fed stood ready to lend any necessary amount of cash to the banks and to open a pipeline to the brokerage houses and investment banks that had lost money. On paper, U.S. investors had lost $500 billion in one day and most of the major U.S. financial institutions had become insolvent overnight with a greater value of outstanding liabilities than assets. The investment banks and brokerage houses were in desperate need of an injection of cash in order to settle short-term debts and remain afloat. The Fed recognized this need and announced that it would provide whatever liquidity was required.

The injection of credit by the Fed was successful: Asset values recovered and a second great depression did not materialize. By July 1989, the S&P 500 had regained its August 1987 peak and, as the market climbed, the financial assets of brokerage houses and investment banks were restored and U.S. financial institutions were able to repay the loans that had kept them solvent. An alternative outcome was always possible. If commercial banks had refused to lend to the brokerage houses and if major investment banks such as Goldman Sachs, Morgan Stanley, or Merrill Lynch had declared bankruptcy, the drop in the value of the market could well have become sustained and self-fulfilling. Then, as now, a great deal depends on the confidence of individual investors.

The 2008 Crash

In October 2008, the S&P 500 lost 40% of its value, and many commentators began to compare the situation to the Great Depression. The experience of the 1987 crash demonstrates that large drops in the value of the stock market are not always accompanied by depressions. So is 2008 more like 1930 or more like 1987? There are parallels with both situations. As in the 1930s, stock market wealth and the value of houses fell and unemployment began to climb. Unlike the 1930s, the Fed responded aggressively by lending money to the major financial institutions.

Consider first the behavior of unemployment and stock market wealth. Figure 9.3 plots the monthly values of the S&P and the unemployment rate from August 1998 to August 2008. The shaded regions represent the dates of the last two recessions, one of which began in March 2001 and ended in November of the same year, and the second of which began in December 2007.

It is clear from this figure that the value of stock market wealth fell dramatically from its peak in February 2007, and as of January 2008, it showed no sign of recovery. The fact that the economy entered a recession in December 2007 suggests that the drop in wealth had already begun to affect the real economy, and this fact alone suggests a closer parallel with the 1930s than with 1987. This is the first disturbing fact about the 2007 recession.

Housing and Stocks: Twin Peaks

A second disturbing fact is that housing prices were declining at the same time as a decline in the stock market. This fact makes the 2007 recession different from the 2001 recession, and according to a historical index of house prices constructed by Robert Shiller, the 2008 fall in nominal house prices was much bigger than the 10.5% drop in 1932, at the worst point of the Great Depression. Figure 9.3 shows not only the current recession but also the recession that lasted from March to November 2001. The 2001 recession occurred shortly after the end of a huge rise in the value of stocks that is popularly referred to as the “dot-com boom,” because it was associated with the creation of companies based around information technology that had little or no current revenues but a small chance of very large potential future earnings. In 1999, the dot-com bubble collapsed and the value of stock market wealth began to decline.

The 2000 stock market collapse was different from the 2008 collapse because in 2000, stock market wealth was replaced by housing equity. At the same time that the stock market declined, the United States entered into a housing boom, and although U.S. households lost money from their stock portfolios, they gained from housing. It is likely that the housing boom helped households to maintain consumption, because as stock market wealth declined, housing wealth increased. This point is illustrated graphically in Figure 9.4, which plots the S&P 500 and an index of housing prices, the Case-Shiller index, from August 1998 through December The Case-Shiller index is measured as the solid line and is plotted on the right axis. The S&P stock market index is measured as the line with circles and is plotted on the left axis. As the stock market began to decline in 2000, the value of house prices climbed to offset the fall in stocks.

Deregulation and Accounting Rules

Before the Bank Act of 1933, commercial banks in the United States borrowed money from households in the form of bank deposits, and they lent money to other households to invest in houses and durable goods such as cars and refrigerators and to firms and small businesses, which used the money to finance their investment in factories and machines. These loans were illiquid, because although most loans will eventually be repaid, the bank could not ask for them to be repaid immediately.

Because banks made illiquid loans, they typically had much less cash on hand than they needed to meet their liabilities in the form of deposits. If all depositors were to try to withdraw their money immediately, there would not be enough cash to go around. For this reason, banks were often subject to panics. A rumor would spread that a bank had made risky investments and was insolvent, and this would cause all depositors to try to withdraw their money at once. For this reason, the house passed the Bank Act of 1933, sponsored by Senator Carter Glass and Congressman Henry B. Steagall.

The End of Glass-Steagall

The Glass-Steagall Act created the Federal Deposit Insurance Corporation, which guaranteed bank deposits up to $100,000 (recently raised temporarily to $250,000), and it put a wall between commercial banks and investment banks by limiting the riskiness of the assets held by any bank that took direct deposits from the public. Before the GlassSteagall Act, there were no restrictions on the kinds of loans that a bank could make. After the Glass-Steagall Act, banks were separated into two kinds: commercial banks and investment banks. Commercial banks were allowed to invest only in safe assets such as government bonds, and the depositors at these banks were insured by the federal government. Investment banks continued to make risky loans, but the owners of the liabilities of these banks were not insured against loss.

The argument for separating commercial and investment banks is that if the deposits of the bank are insured, then it has an incentive to make risky loans at high interest rates. If the loans perform well and are repaid, then the owners of the bank will make high profits. If the loans fail, then the depositors of the banks are protected by the government. Because depositors know that their savings are protected, they will be willing to place their money with the bank without risk and the taxpayers will be left to foot the bill if the loans turn sour.

Many of the provisions of the Glass-Steagall Act were repealed by the Gramm-Leach-Bliley Act of 1999. The Gramm-Leach-Bliley Act was followed by the movement of commercial banks into riskier high-return lending and was one of the main contributing factors to the trouble that now faces the financial services industry. Citibank and Bank of America were free to make riskier investments—and they did!

The repeal of the Glass-Steagall Act caused a reform of the banking industry that left all banks that took depositors’ money free to invest that money anywhere. Throughout the United States, banks that had previously been purely commercial began to move into investment banking by purchasing riskier portfolios of assets including mortgagebacked securities. These securities were one of the main triggers of the financial meltdown that led to the collapse of Lehman Brothers in September 2008.

Was Deregulation to Blame?

Deregulation has been widely criticized as the main cause of the 2008 financial crisis. Although it is certainly a contributing factor, I do not believe it is the main cause. There are two main criticisms of the regulatory changes that occurred in the 1990s. The first is that the repeal of the Glass-Steagall separation of commercial and investment banking led commercial banks to take unnecessary risks with depositors’ funds. This overstates the case. The repeal of Glass-Steagall simply codified an implicit guarantee that had been there all along.

During the 1987 financial crisis, it was the investment banks that were in trouble—not the commercial banks—and at the time, the Glass-Steagall Act was still in place. Nevertheless, Alan Greenspan, implicitly or explicitly, channeled cash to the investment banks to prevent their collapse. Were these banks undergoing a temporary liquidity crisis or were they insolvent? The answer to this question is important but by no means obvious, and it is a question to which classical and Keynesian economics give different answers.

Illiquidity or Insolvency?

Is the banking sector insolvent? If the classical economists are right, then the answer to this question does not depend on the actions of the Fed. In 1987, the investment banks were either insolvent or they were not. If the stock market crash had been caused by investors correctly anticipating that future fundamentals were bad, then the banks were insolvent and Greenspan, by bailing them out, was wasting taxpayers’ money. The fact that the bailout was successful implies, according to this view, that Greenspan knew more than the markets. He is indeed a wizard, and he was able to correctly forecast that the fundamentals of the economy were in fact sound.

If Keynesian economics is right, there was a real danger in 1987 that the drop in the market would turn out to be self-fulfilling. By pumping liquidity into the system, Greenspan convinced investors to regain confidence in stocks and his actions rescued the economy from a second great depression that could have occurred but was not inevitable. This was a gamble that paid off then, but it is not guaranteed always to have the same outcome. If the actions of the Fed had not restored confidence, the Fed would have lost a lot of taxpayer money and a depression would have occurred in spite of its actions.

What Will Happen Next?

Is the 2008 situation more like 1929 or more like 1987? There are parallels with both cases. Like 1987, the Fed stepped in to provide liquidity to the markets. But as of July 2009, the markets had not been noticeably calmed. For most of the postwar period, house prices had gone up and housing was widely viewed as a relatively safe investment. The strong performance of the housing market led to the creation of mortgage-backed securities, which are assets whose payments of principal and interest derive from cash flows generated by mortgages. This market was nonexistent in 1970 but had grown to $7.5 trillion by the end of 2007, of which roughly $6 trillion was issued by the semiautonomous bodies Ginnie Mae and Freddie Mac and implicitly guaranteed by the U.S. government. It is the portion of the mortgage-backed-securities market that is not guaranteed that triggered the financial crisis of 2008.

When U.S. house prices began to fall in 2007, the default rates on mortgage-backed securities turned out to be higher than anyone had predicted, and financial institutions throughout the world discovered that they were holding assets that were potentially worth much less than their book value. But what were they really worth? This question began as one of confidence in the U.S. economy and it spreadin January 2009—to one of the confidence of investors worldwide. Factories, machines, and houses, worldwide, are worth what investors will pay for them. If these assets remain undervalued, then default rates on mortgages in the United States and elsewhere will be high and a catastrophic drop in property values will be self-fulfilling. As wealth drops, demand will fall and workers will lose their jobs all over the world. It would be a grave mistake to think that a worldwide depression of this kind cannot happen again. It would be an equally grave mistake to assume that a depression, if it occurs, is inevitable and is caused purely by fundamentals that are beyond our control. But is there anything that government can or should do about the situation? I will take up that question in the following chapters.

Traditional Monetary Policy

When the economy moves into recession, the central bank lowers the yield on safe assets. In the United States, this works through Fed purchases of Treasury bills on the open market. In the UK and in Europe, the mechanism is a little different but the effect is the same. The central bank lowers the interest rate to stimulate the economy in a recession by injecting liquidity into the financial system. As commercial banks try to lend the additional money, interest rates fall and more risky business ventures become profitable. As firms and households begin to purchase more goods, this increases employment and helps to bring the economy out of recession. This strategy has worked in all of the last 10 postwar recessions, but it is not available to the Fed, the European Central Bank, or the Bank of England today.

There have been 10 complete recessions in the United States since the end of World War II and in every one of them the Fed helped the economy out of a recession by lowering the Fed funds rate. But today, just as in 1934, the interest rate on Treasuries is close to zero, and traditional monetary policy can’t go any further. If the Fed buys Treasury bills and replaces them with cash, it is replacing one zero interest rate government liability with another. This is a little like exchanging a 10-dollar bill for two fives. When the interest rate on Treasury bills is zero, money and bonds become perfect substitutes and the traditional method of running open market operations is like “pushing on a string”.

Quantitative Easing

Since the scope for traditional monetary policy is limited, the Fed and the Bank of England are following a new approach: quantitative easing. The European Central Bank is following a similar strategy, and all three are injecting money into the system in an effort to bring down interest rates on risky assets and long-term government debt. Traditionally, the Fed has conducted open market operations by buying and selling three-month Treasury bills. Quantitative easing refers to an alternative monetary policy of expanding the money supply by buying a range of alternative assets including corporate debt and long-term government bonds.

In August 2008, the Fed owned assets of approximately $800 billion, most of it in the form of three-month Treasury bills. By January 2009, that figure had more than doubled as the Fed made short-term loans to the financial sector and began direct interventions in the commercial paper market. Commercial paper is an unsecured loan with a duration of less than nine months, and in normal times these loans are issued by commercial banks to nonfinancial corporations that use the funds to meet payroll and short-term obligations.

Since November 2008, the Fed has begun to buy commercial paper and has replaced commercial banks as the main lender in this market. By directly lending in the commercial paper market, the Fed is engaging in quantitative easing. Chairman Bernanke has indicated that we are likely to see much more of this in the future, and the Bank of England’s monetary policy committee is committed to a similar approach.

Bernanke’s Plan

Bernanke pointed out that although the interest rate on short-term securities may be close to zero, it is still possible for the nation’s central bank to purchase other kinds of assets. One example is the purchase of long-term government bonds. This policy would work in exactly the same way as traditional monetary policy, but instead of buying three-month Treasury bills—for which the interest rate is already at zero—the Fed would buy longer-term government bonds. As of January 2009, long-term government bonds were yielding a return of between 1.5% and 2.0% depending on the maturity. The hope was that, as government drives down the return on long-term bonds, households will put their money back into the stock market, and U.S. corporations will be encouraged to invest in new plants and equipment.

In his 2003 address in Japan, Chairman Bernanke laid out a program by which the Treasury and the Fed would cooperate to coordinate monetary and fiscal policy. If his proposal were to be applied to the U.S. situation, it would involve the following steps. First, the Treasury would announce a large fiscal stimulus to be paid for by printing money. Second, the Fed would announce an explicit target for the inflation rate. The purpose of this announcement is to anchor the expectations of the public by giving them a clear guide to Fed intentions in future months if inflation begins to reappear. Third, the Fed would engage in a program of quantitative easing by buying a range of assets other than the traditional Fed purchases of three-month Treasury bills.

As of January 2009, the Obama administration had announced plans for an $800 billion increase in the federal deficit and, if Bernanke were to follow step one of the strategy outlined in his advice to the Japanese, the Fed would finance the Obama plan by printing money. Under step two of the plan, as the economy began to expand, the Fed would sell government securities on the open market to absorb the excess liquidity that it created in step one. This would prevent the money supply from expanding too quickly and generating inflation. Step three of the Bernanke planquantitative easing—would involve the purchase of a range of alternative private and government securities including corporate paper and long-term bonds issued by government and private corporations.

How would the Bernanke plan work? The following section explains what the Fed, the Bank of England, and the European Central Bank believe and why, in the view of central bankers, a plan like this makes sense.

What Central Bankers Think

Central bank policymakers throughout the world believe there is a natural rate of unemployment that the economy moves toward in the long run and they believe that monetary and fiscal policy cannot influence this natural rate. In the short run, however, they believe that the unemployment rate may move away from its natural rate because of shocks to demand and supply. An example of a supply shock is a big unexpected increase in the price of oil, such as occurred in 1973 and again in 1979. An example of a demand shock is a big unexpected drop in house prices of the kind we saw in 2007.

When The Bubble Bursts

The consensus central bank position is that demand shocks, such as a bursting bubble, can affect real activity only to the extent that prices are slow to adjust. It follows that there will need to be a period of price adjustment as house prices fall to a new lower, sustainable level. That adjustment process will likely require money wages and the prices of many other commodities to fall, too. Because wages and prices cannot fall fast enough, unemployment will increase in the short term.

From the central bank perspective, there is an alternative preferable scenario to a protracted period of falling prices. It is to flood the economy with money so that prices and wages do not have to fall. Monetary policy stimulates demand by lowering interest rates, but this cannot work today, because the interest rate is already at zero. Quantitative easing involves trying to bring down the spreads on risky assets by lending directly to firms or by bringing down the premium on long bonds by directly purchasing longterm government debt. This might work, but there is no guarantee that it will cause firms to invest in new factories and machines. Instead, we might end up in a situation where corporate paper and long bonds pay a zero interest rate and the Fed absorbs any associated default risk.

Learning from the Great Depression

The next best thing to holding everything constant is to make a very large change in a variable of interest and see what happens to everything else. For example, if we observe an unusually large increase in government expenditure, then we can infer that unusual simultaneous changes in employment, consumption, or GDP were probably caused by the change in government purchases.

Large dramatic increases in the size of government don’t occur often. Even during the Great Depression, the increase in government in the United States was gradual and the really dramatic changes did not occur until World War II. At that time, however, we did see a remarkable and unusual change as government purchases on goods and services increased from 12% of GDP in 1940 to 50% in 1944: It was the wartime stimulus that pulled the United States out of the Great Depression, not Roosevelt’s New Deal, as some commentators have claimed.

Stages of Recovery

A couple of things stand out on this figure. First, in stage 1 of the recovery from the Great Depression, between 1933 and 1937, the real value of government expenditures was tiny. Observe the lower line in Figure 10.1, which looks flat between 1933 and 1937. The size of Roosevelt’s stimulus was negligible compared with the later wartime expansion and the increase in government purchases over this period looks a lot more like a slowly moving gentle trend than a Keynesian fiscal experiment.

Second, in stage 2 of the recovery, from 1938 to 1942, the increase in the size of government was truly massive. In Figure 10.1, this shows up as the upward blip in government purchases (the lower line in the figure) during World War II. Large movements of this kind are rare in economic time series, but they are incredibly valuable because large unpredicted movements are natural experiments. We can use this experiment to ask: How large was the multiplier during World War II?

The data from World War II suggest that, in the current economic climate, an $800 billion increase in government purchases in the United States is likely to lead to an $800 billion increase in GDP, or at best two-thirds of the increase that the administration is projecting since their calculations are based on the assumption that the multiplier is 1.5. Although the proposed Obama stimulus package will create jobs, it is unlikely that they will be created in the private sector, and it seems that we may be headed for the largest increase in the size of government since World War II.

Two Reasons For Government Deficits

There are two reasons for running budget deficits. The first is that unemployment is too high and social resources are being wasted. The second is to fund projects that government can provide more effectively than the private sector. These two reasons are being confounded in the current debate, and I am concerned that the possibility of the world entering a second Great Depression is being used, at least in the United States, as a tactic to rush new spending programs into law without proper debate.

Do We Need A Bigger Goverment?

The best way to swiftly enact a fiscal stimulus is to print money and hand out checks to every taxpayer. A rebate of $2,700 per household seems like a good place to start in the U.S. case. As the economy begins to recover, the Fed should remove the excess liquidity from the system through an announced policy of raising the interest rate to meet an inflation target of 2%. This would help to reduce the threat that the stimulus will create inflation. The expenditure component of the Obama proposal, in my view, requires more considered debate and should be treated for what it is: a long-term investment in America’s future.

Not all infrastructure projects are worthwhile. The Alaska bridge to nowhere that was a focus of the recent political campaign is a case in point, and it is because some social investments are ill considered that national legislatures should carefully weigh the costs and benefits of each project before committing taxpayer money to possible wasteful spending.

Give Me a One-Armed Economist

Here are my views on fiscal policy. A large fiscal stimulus may or may not be an important component of a recovery plan. I believe there is a better alternative to fiscal policy, which I explain in chapter 11. But if a fiscal policy is used, it should take the form of transfer payment to every domestic resident, not an increase in government expenditure.

Here are my views on monetary policy. Short-term interest rates should be increased as soon as feasible, because a positive interest rate is needed if a national central bank is to effectively control inflation. In the future, central banks should use the interest rate for this purpose and not to prevent recessions.

But if a central bank raises the domestic interest rate without independently managing confidence, the result will be a drop in the value of the national stock market and a further deterioration in the real economy. To prevent this from happening, central banks need a second instrument. The following chapter explains what this instrument is and how to use it. I propose a strategy for managing confidence that will put an end to the cycle of boom and bust that has characterized capitalist economies for the past 300 years.

What Happened in 2008

It is widely believed by economists and financial journalists that we were in a liquidity crisis in 2008. The talk in Washington, Frankfurt, Tokyo, and London was that we must unfreeze the credit markets, and many observers expressed puzzlement that investment banks and other financial institutions were holding onto capital, newly acquired from national central banks, and were refusing to lend it to corporations. But in fact, the strategy of holding onto cash was rational for each bank individually.

Banks in 2009 were concerned that the market had further to fall. If the values of houses, factories, and machines were to continue to fall, loans to corporations would not be repaid and the banks would lose money. As of February 2009, the market was in a holding pattern in which market participants were terrified that things could get much worse.

Adding a New Policy Lever

A nation’s central bank controls two levers of economic policy. The first is the size of its balance sheet, which determines the stock of money circulating in the economy. The second is the composition of its balance sheet. Until recently, central banks have almost exclusively held reserves in the form of gold or more recently as loans to national governments. With the advent of inflation targeting in the 1980s, central bankers learned to use the first lever to control inflation, but only now are they experimenting with the second. It’s a bit like learning how to turn the rudder of a sailboat without knowing how to raise or lower the sails.

In November 2008, the Fed and the Bank of England began to purchase different kinds of assets in a policy of quantitative easing. This was the right approach, but it didn’t go far enough. It is time for an expanded role for direct policy intervention in the asset markets through a policy that targets not just one price, the short-term interest rate, but two. In addition to using the size of their balance sheets to change the interest rate and target inflation, policymakers should use the composition of central bank balance sheets to change asset prices and target the unemployment rate. This need not involve the direct purchase of voting shares in individual companies. A better plan would be to trade a stock market index fund.

My Argument Summarized

Free market economies do not provide the necessary price signals to ensure that a given number of jobs is filled in the right way. Because these signals are missing, a market economy can become stuck with very high unemployment that doesn’t go away.

Firms decide how many workers to hire based on the demand for the goods that they produce. The demand for goods depends on wealth, and different levels of wealth lead to different unemployment rates. The wealth of households depends on what other households believe. Wealth depends on confidence! For every self-fulfilling belief about the value of wealth there is an unemployment rate that will persist for a very long time, and each of these unemployment rates is associated with a different set of prices for houses and a different set of prices for factories and machines. The value of physical assets depends on what market participants think they will be worth in the future.

Fiscal and monetary policy cannot help us to escape from high unemployment unless these policies also restore confidence in the stock market. There is no guarantee that this will happen. In the summer of 2009, many economists were predicting that economic growth would reappear in the United States and in the UK in the third or fourth quarter of the year and that the economy would begin to recover. Nobody was predicting that unemployment would fall any time soon.

It is my view that when world economies emerge from the 2008 crisis, forecasters in every country in the world will revise upward their estimates of their nations’ natural rates of unemployment in response to the fact that the world recession has permanently destroyed jobs. Central bankers will be forced to raise domestic interest rates while unemployment is still well above its prerecession rate in order to prevent inflation.

From the perspective of classical and new-Keynesian theory, central bankers will be making the correct decision. They believe that the new higher rates of unemployment that will emerge are permanent features of their domestic economies that are due to changes in economic fundamentals. In a sense, they are right; but the fundamental that has changed is the confidence of stock market participants, and confidence is not independent of policy.

The correct response to the crisis is to set in place, in every country in the world, an institution to control the value of national stock market wealth by targeting the rate of growth of an index fund. Ideally, this function would be taken on by the nation’s central bank and coordinated with domestic monetary policy. Central banks should use changes in the size of their balance sheets to prevent inflation from rising too high or too low. They should use changes in the composition of their balance sheets to prevent bubbles and crashes.

End.