Debunking Economics Part 19

The Financial Instability Hypothesis Minsky's starting point was that, since the Great Depression had occurred, and since similar if smaller crises were a recurrent feature of the nineteenth century, before 'Big Government' became the norm in market economies, an economic model had to be able to generate a depression as one of its possible outcomes: 'Can "It" a Great Depression happen again? And if "It" can happen, why didn't "It" occur in the years since World War II? These are questions that naturally follow from both the historical record and the comparative success of the past thirty-five years. To answer these questions it is necessary to have an economic theory which makes great depressions one of the possible states in which our type of capitalist economy can find itself' (Minsky 1982: 5; emphasis added).

For this reason, Minsky explicitly rejected neocla.s.sical economics: The abstract model of the neocla.s.sical synthesis cannot generate instability. When the neocla.s.sical synthesis is constructed, capital a.s.sets, financing arrangements that center around banks and money creation, constraints imposed by liabilities, and the problems a.s.sociated with knowledge about uncertain futures are all a.s.sumed away. For economists and policy-makers to do better we have to abandon the neocla.s.sical synthesis. (Ibid.: xiii) In place of the non-monetary, equilibrium-fixated, uncertainty-free, inst.i.tutionally barren and hyper-rational individual-based reductionist neocla.s.sical model, Minsky's vision of capitalism was strictly monetary, inherently cyclical, embedded in time with a fundamentally unknowable future, inst.i.tution-rich and holistic, and considered the interactions of its four defining social ent.i.ties: industrial capitalists, bankers, workers and the government.

I published my first paper on Minsky's hypothesis in 1995 (Keen 1995), and the following summary of Minsky's verbal model of a financially driven business cycle is reproduced from that paper.2 I provide it verbatim here since its conclusion written in 1993, long before neocla.s.sical economists began to congratulate themselves about the 'Great Moderation' shows that the calamity the world economy fell into in 2007/08 was not an unpredictable 'Black Swan' event, but something that was entirely foreseeable with the right economic theory: Minsky's a.n.a.lysis of a financial cycle begins at a time when the economy is doing well (the rate of economic growth equals or exceeds that needed to reduce unemployment), but firms are conservative in their portfolio management (debt to equity ratios are low and profit to interest cover is high), and this conservatism is shared by banks, who are only willing to fund cash-flow shortfalls or low-risk investments. The cause of this high and universally practiced risk aversion is the memory of a not too distant system-wide financial failure, when many investment projects foundered, many firms could not finance their borrowings, and many banks had to write off bad debts. Because of this recent experience, both sides of the borrowing relations.h.i.+p prefer extremely conservative estimates of prospective cash flows: their risk premiums are very high.

However, the combination of a growing economy and conservatively financed investment means that most projects succeed. Two things gradually become evident to managers and bankers: 'Existing debts are easily validated and units that were heavily in debt prospered: it pays to lever' (Minsky 1982, p. 65). As a result, both managers and bankers come to regard the previously accepted risk premium as excessive. Investment projects are evaluated using less conservative estimates of prospective cash flows, so that with these rising expectations go rising investment and a.s.set prices. The general decline in risk aversion thus sets off both growth in investment and exponential growth in the price level of a.s.sets, which is the foundation of both the boom and its eventual collapse.

More external finance is needed to fund the increased level of investment and the speculative purchase of a.s.sets, and these external funds are forthcoming because the banking sector shares the increased optimism of investors (Minsky 1980, p. 121). The accepted debt to equity ratio rises, liquidity decreases, and the growth of credit accelerates.

This marks the beginning of what Minsky calls 'the euphoric economy' (Minsky 1982, pp. 120124), where both lenders and borrowers believe that the future is a.s.sured, and therefore that most investments will succeed. a.s.set prices are revalued upward as previous valuations are perceived to be based on mistakenly conservative grounds. Highly liquid, low-yielding financial instruments are devalued, leading to a rise in the interest rates offered by them as their purveyors fight to retain market share.

Financial inst.i.tutions now accept liability structures for both themselves and their customers 'that, in a more sober expectational climate, they would have rejected' (Minsky 1980, p. 123). The liquidity of firms is simultaneously reduced by the rise in debt to equity ratios, making firms more susceptible to increased interest rates. The general decrease in liquidity and the rise in interest paid on highly liquid instruments triggers a market-based increase in the interest rate, even without any attempt by monetary authorities to control the boom. However, the increased cost of credit does little to temper the boom, since antic.i.p.ated yields from speculative investments normally far exceed prevailing interest rates, leading to a decline in the elasticity of demand for credit with respect to interest rates.

The condition of euphoria also permits the development of an important actor in Minsky's drama, the Ponzi financier (Minsky 1982, pp. 70, 115 [...]). These capitalists profit by trading a.s.sets on a rising market, and incur significant debt in the process. The servicing costs for Ponzi debtors exceed the cash flows of the businesses they own, but the capital appreciation they antic.i.p.ate far exceeds the interest bill. They therefore play an important role in pus.h.i.+ng up the market interest rate, and an equally important role in increasing the fragility of the system to a reversal in the growth of a.s.set values.

Rising interest rates and increasing debt to equity ratios eventually affect the viability of many business activities, reducing the interest rate cover, turning projects that were originally conservatively funded into speculative ones, and making ones that were speculative 'Ponzi.' Such businesses will find themselves having to sell a.s.sets to finance their debt servicing and this entry of new sellers into the market for a.s.sets p.r.i.c.ks the exponential growth of a.s.set prices. With the price boom checked, Ponzi financiers now find themselves with a.s.sets that can no longer be traded at a profit, and levels of debt that cannot be serviced from the cash flows of the businesses they now control. Banks that financed these a.s.sets purchases now find that their leading customers can no longer pay their debts and this realization leads initially to a further bank-driven increase in interest rates. Liquidity is suddenly much more highly prized; holders of illiquid a.s.sets attempt to sell them in return for liquidity. The a.s.set market becomes flooded and the euphoria becomes a panic, the boom becomes a slump.

As the boom collapses, the fundamental problem facing the economy is one of excessive divergence between the debts incurred to purchase a.s.sets, and the cash flows generated by them with those cash flows depending upon both the level of investment and the rate of inflation.

The level of investment has collapsed in the aftermath of the boom, leaving only two forces that can bring a.s.set prices and cash flows back into harmony: a.s.set price deflation, or current price inflation. This dilemma is the foundation of Minsky's iconoclastic perception of the role of inflation, and his explanation for the stagflation of the 1970s and early 1980s.

Minsky argues that if the rate of inflation is high at the time of the crisis, then though the collapse of the boom causes investment to slump and economic growth to falter, rising cash flows rapidly enable the repayment of debt incurred during the boom. The economy can thus emerge from the crisis with diminished growth and high inflation, but few bankruptcies and a sustained decrease in liquidity. Thus, though this course involves the twin 'bads' of inflation and initially low growth, it is a self-correcting mechanism in that a prolonged slump is avoided.

However, the conditions are soon reestablished for the cycle to repeat itself, and the avoidance of a true calamity is likely to lead to a secular decrease in liquidity preference.

If the rate of inflation is low at the time of the crisis, then cash flows will remain inadequate relative to the debt structures in place. Firms whose interest bills exceed their cash flows will be forced to undertake extreme measures: they will have to sell a.s.sets, attempt to increase their cash flows (at the expense of their compet.i.tors) by cutting their margins, or go bankrupt. In contrast to the inflationary course, all three cla.s.ses of action tend to further depress the current price level, thus at least partially exacerbating the original imbalance. The a.s.set price deflation route is, therefore, not self-correcting but rather self-reinforcing, and is Minsky's explanation of a depression.

The above sketch basically describes Minsky's perception of an economy in the absence of a government sector. With big government, the picture changes in two ways, because of fiscal deficits and Reserve Bank interventions. With a developed social security system, the collapse in cash flows that occurs when a boom becomes a panic will be at least partly ameliorated by a rise in government spending the cla.s.sic 'automatic stabilizers,' though this time seen in a more monetary light. The collapse in credit can also be tempered or even reversed by rapid action by the Reserve Bank to increase liquidity. With both these forces operating in all Western economies since World War II, Minsky expected the conventional cycle to be marked by 'chronic and ... accelerating inflation' (Minsky 1982, p. 85). However, by the end of the 1980s, the cost pressures that coincided with the slump of the early 1970s had long since been eliminated, by fifteen years of high unemployment and the diminution of OPEC's cartel power. The crisis of the late 1980s thus occurred in a milieu of low inflation, raising the specter of a debt deflation. (Keen 1995: 61114) I added the following qualification about the capacity for government action to attenuate the severity of a debt deflation while not addressing its underlying causes to my precis of Minsky in the first edition of Debunking Economics: If a crisis does occur after the Internet Bubble finally bursts, then it could occur in a milieu of low inflation (unless oil price pressures lead to an inflationary spiral). Firms are likely to react to this crisis by dropping their margins in an attempt to move stock, or to hang on to market share at the expense of their compet.i.tors. This behavior could well turn low inflation into deflation.

The possibility therefore exists that America could once again be afflicted with a debt deflation though its severity could be attenuated by the inevitable increase in government spending that such a crisis would trigger. America could well join j.a.pan on the list of the global economy's 'walking wounded' mired in a debt-induced recession, with static or falling prices and a seemingly intractable burden of private debt. (Keen 2001a: 254) That a crisis might occur, and even that government action might attenuate it, was something that one could antic.i.p.ate with Minsky's verbal economic theory. But a market economy is a complex system the most complex social system that has ever existed and its very complexity means that feedback effects might occur that are simply impossible to predict with a verbal model alone. For that reason, in my PhD I decided to attempt what Minsky had not succeeded in doing: to provide a mathematical model that did justice to the compelling verbal description he gave of debt deflation.

Modeling Minsky Minsky did develop a mathematical model of a financially driven business cycle in his PhD, which resulted in the one paper he ever had published in a mainstream economic journal, the American Economic Review (Minsky 1957).3 But the model was unsatisfactory for a number of reasons, and he subsequently abandoned it to stick with predominantly verbal reasoning.

Minsky's failure to develop a satisfactory mathematical model was partly due to bad timing: the 1950s pre-dated the development of complexity theory, which made trying to build a model of his hypothesis virtually impossible. Minsky simply added a financial dimension to the dominant linear trade cycle model of the day, which was a particularly unsuitable foundation for his hypothesis.4 In 1993, well after complexity theory had developed, I built my initial Minsky model using the far more suitable foundation of the cyclical growth model developed by the non-neocla.s.sical economist Richard Goodwin (Goodwin 1967).

Goodwin's model considered the level of investment and the distribution of income in a simple two-cla.s.s model of capitalism. A high initial wage and high rate of employment meant that wages absorbed most of output, so that profit was low and therefore investment was low. The low rate of investment meant that the capital stock grew slowly (or fell because of depreciation), leading to a low rate of growth of output (or even falling output) and hence a growing unemployment rate since population growth would then exceed the rate of economic growth.

The rising unemployment rate reduced workers' bargaining power, leading to stagnant or falling wages which increased capitalists' profit share. They then increased investment, leading to a boom that drove the employment rate up, which strengthened the bargaining power of workers. Wages then rose and, because employment was high, wages absorbed most of output which is where the cycle began.5 This was a cla.s.sic dynamic model of 'circular causation' that is very common in biological modeling, but sadly a rarity in economics because of the neocla.s.sical obsession with equilibrium. It also had a startling characteristic compared to the standard fare in economics: it was inherently cyclical. Given an arbitrary starting point, the model generated regular cycles in both the distribution of income and the employment rate. There was no tendency toward equilibrium, but no tendency to breakdown either: the same cycle repeated for ever.

13.1 Goodwin's growth cycle model Economists were falsely of the opinion that this was impossible. As John Hicks (remember him?) put it: 'A mathematically unstable system does not fluctuate; it just breaks down. The unstable position is one in which it will not tend to remain' (Hicks 1949).

As is so often the case, Hicks was right in particular and wrong in general. If they were unstable, then dynamic versions of the linear models that he and most neocla.s.sical economists worked with would indeed break down by returning impossible values for variables, such as negative prices or infinite levels of output. But Goodwin's model was inherently nonlinear, because two variables in the system the wage rate and the level of employment had to be multiplied together to work out wages and hence profits. As I explained in Chapter 9, nonlinear models can have persistent cycles without breaking down.

The professor of applied mathematics turned non-orthodox economist John Blatt observed that Goodwin's model was the best of the many dynamic economic models he had reviewed, and suggested that it would provide an excellent foundation for modeling financial dynamics in capitalism. In stark contrast to the neocla.s.sical obsession with equilibrium, one of Blatt's criticisms of Goodwin's basic model was that its equilibrium was not unstable: Of course, the model is far from perfect. In particular, we feel that the existence of an equilibrium which is not unstable (it is neutral) is a flaw in this model [...] The first flaw can be remedied in several ways [...] [such as] introduction of a financial sector, including money and credit as well as some index of business confidence. Either or both of these changes is likely to make the equilibrium point locally unstable, as is desirable [...] But, while it is obvious that much work remains to be done, we have no doubt that the Goodwin model is the most promising of all the 'schematic models' of the trade cycle and well deserves further investigation. (Blatt 1983: 21011) I took up Blatt's suggestion in my PhD, by adding Keynes's model of how capitalists form conventions to cope with uncertainty, and Minsky's emphasis upon the role of debt in financing investment plans during a boom.

Of Keynes's three conventions to cope with uncertainty, the most important was the tendency to project forward current conditions: 'We a.s.sume that the present is a much more serviceable guide to the future than a candid examination of past experience would show it to have been hitherto' (Keynes 1937: 214).

A simple way to capture this in a mathematical model was to argue that capitalists would invest very little when the rate of profit today was very low, and invest a lot when the rate of profit was high. This was easily captured by replacing Goodwin's simple but unrealistic a.s.sumption that capitalists invested all their profits with a nonlinear relations.h.i.+p that meant investment would be less than profits when the rate of profit was low, and more than profits when the rate of profit was high.

Minsky improved upon Keynes by incorporating the insights of Schumpeter and Fisher on the essential role of debt in a capitalist economy: when capitalists' desire to invest exceeded retained earnings as they would do during a boom then capitalists would borrow to finance the additional investment. I introduced this with a simple differential equation that said the rate of change of debt equaled investment minus profits.6 My first Minsky model This added one additional dynamic to Goodwin's model: the rate of change of debt, which rose when investment exceeded profits and fell when profits exceeded investment. During a boom, capitalists borrow to finance investment, and this drives up the debt-to-output ratio. During a slump, capitalists invest less than profits, and this reduces the debt-to-output ratio. The change in the debt ratio then affects the rate of profit, since profits are now equal to output, minus wages, minus interest on outstanding debt.

This simple extension to Goodwin's model dramatically altered its behavior. Goodwin's basic model generated fixed cycles indefinitely; this extended system could generate several different outcomes, ranging from a convergence to equilibrium values for income distribution, the employment rate and the debt-to-output ratio; cycles in all three variables of varying magnitudes over time; or a blowout in the debt-to-GDP ratio: a debt-induced depression.

The model also had three fascinating and, as it turned out, prescient characteristics.

First, even though capitalists were the only borrowers in this simple model, the debt repayment burden actually fell on workers: the wages share of output fell as the debt level rose, while the profit share fluctuated around an equilibrium value.

Secondly, if the model did head toward a debt-induced breakdown, the debt-to-output ratio ratcheted up over time: debt would rise during a boom, reach a peak and then fall during a slump, but a new boom would begin before the debt-to-output ratio had dropped to its original value.

13.2 My 1995 Minsky model Thirdly, the breakdown was preceded by a period of reduced volatility: fluctuations in employment and output would start off very large and then fall the model generated a 'Great Moderation' before one appeared in the empirical record. But slowly, as the debt ratio rose even higher, the volatility started to rise again, until there was one last extreme cycle in which the debt level went so high that debt repayments overwhelmed the capacity of capitalists to pay.

13.3 The vortex of debt in my 1995 Minsky model The economy then went into a death spiral as the level of debt overwhelmed the capacity of capitalists to service that debt. A 'Great Moderation' gave way to a 'Great Recession' see Figure 13.3.

When I first completed this model in April 1992, the 'Great Moderation' had yet to begin, but the peculiar dynamics of the model struck me as remarkable. This led me to finish my first published paper on this model with a flourish that, at the time, seemed grandiose, but which ultimately proved to be prophetic: From the perspective of economic theory and policy, this vision of a capitalist economy with finance requires us to go beyond that habit of mind which Keynes described so well, the excessive reliance on the (stable) recent past as a guide to the future. The chaotic dynamics explored in this paper should warn us against accepting a period of relative tranquility in a capitalist economy as anything other than a lull before the storm. (Keen 1995: 634; emphasis added) However, Minsky had also noted that government spending could stabilize an unstable economy. In that same paper I modeled this possibility by introducing government spending as an effective subsidy to capitalists that grew as unemployment rose and fell as it subsided though workers receive unemployment benefits, the unemployed spend everything they get on consumption, so that corporations are the ultimate recipients of government welfare. Similarly, I modeled government taxation of business as rising as profits rose, and falling when profits fell.

As well as adding a fourth 'system state' to the model the level of net government spending as a proportion of output this modified the definition of profit. It was now output, minus wages, minus interest payments on debt, minus taxes plus the government subsidy.

In the model, the presence of government spending acted as a counterweight to the private sector's tendency to acc.u.mulate debt: a rising subsidy and falling taxes during a slump gave business additional cash flows with which to repay debt during a slump, while rising taxes and a falling subsidy during a boom attenuated the private sector's tendency to acc.u.mulate debt.

The result was a system which was inherently cyclical, but in which the cycles stayed within manageable bounds: there was no systemic breakdown, as there had been in the pure private sector model. It was a pure limit cycle of the kind Blatt thought should be generated by a realistic model (Blatt 1983: 211).

13.4 Cyclical stability with a counter-cyclical government sector Reality, I expected, lay somewhat between these two extremes of a private sector en route to a debt-induced breakdown, and a cyclical system kept within bounds by the 'automatic stabilizers' of government spending and taxation. The government sector modeled in this paper 'held the line' against rising unemployment, whereas in the real world governments had retreated from trying to restrain rising unemployment. I also knew that Ponzi-style behavior had become more dominant in the real world over time something that I had not modeled explicitly, since in my model all borrowing led to productive investment. Also, though the models considered the role of private debt, they were only implicitly monetary, and I could not capture the impact of inflation or deflation upon the economy.

So there were ways in which I did not expect the real world to match my models. I resolved to extend them over time to make them explicitly monetary, to model governments that gradually reduced their role as fiscal stabilizers, to incorporate borrowing for purely speculative reasons and so on but in the immediate aftermath I was distracted from this agenda by the ferocious reaction that neocla.s.sical economists had to the chapter 'Size does matter' in the first edition of Debunking Economics. That dispute consumed my research energies in the four years from 2001 till 2005.

Finally in December 2005, I attempted to leave this argument behind and at long last write the book-length treatment of Minsky's hypothesis that I had first committed to do in 1998.7 When I checked the ratio of private debt to GDP for the first time in over a decade, I quickly realized that a crisis would strike long before my technical book on how such crises came about would be ready.

Reality check, December 2005 The last thing I expected was that the real world would be in worse shape than my models implied, but that's what appeared to be the case in December 2005. While drafting an expert witness report on debt in a predatory lending case, I scribbled before I had checked the data that 'Debt to GDP ratios have been rising exponentially.' I expected that I'd need to attenuate that statement once I checked the data the ratio would have been rising, I thought, though not at an exponential rate.

I vividly remember my stunned reaction when I first plotted the data, at about 1 a.m. on 22 December in Perth, Western Australia. Australia's private debt to GDP level had increased more than fivefold since the mid-1960s, and the rate of increase was clearly exponential and it had a burst super-bubble in the 1980s, similar to the cyclical fluctuations in the debt-to-income ratio generated by my Minsky model.

13.5 Australia's private debt-to-GDP ratio, 19752005 I quickly downloaded the US Flow of Funds data to see whether Australia was unique. Obviously, it wasn't see Figure 13.6. This was, as I expected, a global phenomenon. The US debt ratio was slightly less obviously exponential, but had increased even more than the Australian, and over a longer time period. Similar data could be found for most OECD nations, and especially the Anglo-Saxon countries.

Such an exponential rise in the debt ratio had to break, and when it did the global economy would be thrust into a downturn that would surely be more severe than those of the mid-1970s and early 1990s the last times that the bursting of speculative bubbles had caused serious recessions. There was even the prospect that this would be an 'It' break: a debt-induced downturn so severe that the outcome would be not merely a recession, but a depression.

13.6 US private debt to GDP, 19552005 Someone had to raise the alarm, and I realized that, at least in Australia, I was probably that somebody. I once again put Finance and Economic Breakdown on the backburner, and devoted myself to warning the general public and policy-makers of the impending economic crisis. I began with media interviews, progressed to sending out a 'Debt.w.a.tch' report on debt coinciding with the Reserve Bank of Australia's monthly meetings from November 2006 (Keen 2006),8 and in March 2007 I established the Debt.w.a.tch blog (www.debtdeflation.com/blogs).

Raising the alarm was not enough. I also had to dramatically improve my empirical understanding of the role of debt in a capitalist economy, and extend my Minsky model to cover the issues that I clearly had not paid sufficient attention to in 1995: the impact of Ponzi finance, and the active role of the financial sector in financial crises.

The empirical dynamics of debt The key insight about the role of debt in a capitalist society was provided by Schumpeter: in a growing economy, the increase in debt funds more economic activity than could be funded by the sale of existing goods and services alone: 'in real life total credit must be greater than it could be if there were only fully covered credit. The credit structure projects not only beyond the existing gold basis, but also beyond the existing commodity basis' (Schumpeter 1934: 95, 101).

Aggregate demand in a credit-driven economy is therefore equal to income (GDP) plus the change in debt. This makes aggregate demand far more volatile than it would be if income alone was its source, because while GDP (and the level of acc.u.mulated debt) changes relatively slowly, the change in debt can be sudden and extreme. In addition, if debt levels are already high relative to GDP, then the change in the level of debt can have a substantial impact on demand.

A numeric example ill.u.s.trates this process (see Table 13.1). Consider an economy with a GDP of $1,000 billion that is growing at 10 percent per annum, where this is half due to inflation and half due to real growth, and which has a debt level of $1,250 billion that is growing at 20 percent per annum. Aggregate demand will therefore be $1,250 billion: $1,000 billion from GDP, and $250 billion from the increase in debt (which will rise from $1,250 billion to $1,500 billion over the course of the year).

Imagine that the following year, GDP continues to grow at the same 10 percent rate, but debt growth slows down from 20 percent per annum to 10 percent (the debt-to-GDP ratio will therefore stabilize at 150 percent). Demand from income will be $1,100 billion 10 percent higher than the previous year while demand from additional debt will be $150 billion (10 percent of the $1,500 billion level at the start of the year).

Aggregate demand in this second year will thus be $1,250 billion exactly the same as the year before. However, since inflation is running at 5 percent, this will mean a fall in real output of about 5 percent a serious recession. So just a slowdown in the rate of growth of debt can be enough to trigger a recession. An absolute fall in debt isn't needed to cause problems, though it certainly will make things worse still.

TABLE 13.1 A hypothetical example of the impact of decelerating debt on aggregate demand Schumpeter ignored the role of a.s.set markets in the economy, so that in his model the increase in debt financed investment (and the sale of goods financed consumption). Therefore in his model, aggregate demand equals aggregate supply, but part of aggregate demand is debt-financed. In this example, demand financed by the sale of goods and services purchased $1,000 billion of consumer goods, while $250 billion of investment goods were bought on credit. Twenty percent of aggregate demand therefore came from rising debt.

Two consequences follow from this, of which Schumpeter was fully cognizant.

First, the expansion of credit must come, not from someone's savings being transferred to another person via a loan which is the conventional model of how banks operate but by the banking sector creating new money and credit 'out of nothing': [I]n so far as credit cannot be given out of the results of past enterprise [...] it can only consist of credit means of payment created ad hoc, which can be backed neither by money in the strict sense nor by products already in existence [...]

It provides us with the connection between lending and credit means of payment, and leads us to what I regard as the nature of the credit phenomenon [...] credit is essentially the creation of purchasing power for the purpose of transferring it to the entrepreneur, but not simply the transfer of existing purchasing power. (Ibid.: 1067) The banking sector therefore must have the capacity to create purchasing power an issue I return to in the next chapter.

Secondly, the numerical example given here involves an unsustainable rate of growth of debt in the first year, so that there has to be a slowdown in the rate of growth of debt, which will cause a recession. However, the increased debt also helps create productive capacity for the economy, which can later be used to service the debt. There is thus a limit to the severity of cycles that can result: though excessive debt growth will cause a boom, and the inevitable slowdown in the growth of debt will cause a slump, the economy's capacity to produce is expanded by the growth of debt. Serious adjustments might be needed falling prices, debt write-offs as some firms go bankrupt, and so on but ultimately the economy will be able to reduce debt to manageable levels again, and growth will resume once more.

Minsky extended Schumpeter by considering Ponzi finance as well lending to finance the speculative purchase of existing a.s.sets. Now, as well as aggregate demand being both income plus the change in debt, aggregate supply is both the output of new goods and services and the net turnover of existing a.s.sets. This breaches the virtuous cycle that Schumpeter saw between rising debt and a rising capacity to service that debt, because the money borrowed to buy a.s.sets adds to society's debt level without increasing its productive capacity. Thus when a slump follows a debt-fuelled boom, it is possible that debt servicing will exceed the economy's available cash flows leading to not merely a recession, but a depression.

This Minskian process has been playing out in America ever since the mid-1960s when Minsky first developed his Financial Instability Hypothesis. Minsky himself identified 1966 as the time at which America made the transition from a productive to a Ponzi economy: 'A close examination of experience since World War II shows that the era quite naturally falls into two parts. The first part, which ran for almost twenty years (19481966), was an era of largely tranquil progress. This was followed by an era of increasing turbulence, which has continued until today' (Minsky 1982: xiii).

Minsky's judgment was based largely on his financial interpretation of the US business cycle from that point on: The first serious break in the apparently tranquil progress was the credit crunch of 1966. Then, for the first time in the postwar era, the Federal Reserve intervened as a lender of last resort to refinance inst.i.tutions in this case banks which were experiencing losses in an effort to meet liquidity requirements. The credit crunch was followed by a 'growth' recession, but the expansion of the Vietnam War promptly led to a large federal deficit which facilitated a recovery from the growth recession.

The 1966 episode was characterized by four elements: (1) a disturbance in financial markets that led to lender-of-last-resort intervention by the monetary authorities; (2) a recession (a growth recession in 1966); (3) a sizable increase in the federal deficit; and (4) a recovery followed by an acceleration of inflation that set the stage for the next disturbance. The same four elements can be found in the turbulence of 196970, 197475, 1980, and 1981. (Ibid.: xivxv) 13.7 Aggregate demand in the USA, 19652015 Empirically, the late 1960s also marked the point at which the acc.u.mulated debt of the private sector exceeded 100 percent of GDP. From that point on, the dynamics of debt began to dominate macroeconomic performance in the USA first generating a false prosperity, and then a calamitous collapse when the great debt bubble finally burst (see Figure 13.7).

TABLE 13.2 The actual impact of decelerating debt on aggregate demand Note: 1 The change in real demand was the same as the change in nominal demand since inflation was effectively zero in 2009 For the first time since the Great Depression, the aggregate level of private debt began to fall in January 2009. But the economic downturn began well before, when the rate of growth of debt slowed from its peak level, just as the numerical example ill.u.s.trates.

13.8 US private debt The debt bubble went out with a bang: the increase in private sector debt in 2008, the final year of the bubble, was a truly stupendous $4 trillion, which boosted aggregate demand from GDP alone by over 28 percent. A year later, debt was growing by 'only' $1.5 trillion, with the result that aggregate demand slipped from its peak level of US$18.3 trillion in 2008 to $15.7 trillion at the beginning of 2009. Though GDP had fallen slightly over calendar year 2009 from $14.3 trillion to $14.2 trillion by far the biggest hit to the USA's solar plexus came simply from a slowdown in the rate of growth of debt. Though real GDP fell by a mere 2.7 percent, aggregate demand fell by a ma.s.sive 14.2 percent see Table 13.2.

The year 2008 thus brought to a close a period of literally half a century in which private debt had always been growing, and thus adding to aggregate demand. This of itself was not inherently a problem: as both Schumpeter and Minsky argued, rising debt is necessary to finance entrepreneurial activity and to enable the economy to grow. The problem for America, and most of the OECD, was that this increase in debt was rising relative to GDP indicating that what was being funded was not good, Schumpeterian innovation, but bad Ponzi-finance speculation. The annual increase in debt, which had hovered around 5 percent of GDP in the 1950s and 1960s, rose in a series of peaks and troughs to the 28 percent peak of 2008, from where it plunged to a maximum rate of decline of over 18 percent in early 2010 see Figure 13.9.

The $2.6 trillion drop in aggregate demand hit America's a.s.set markets hard. Though the Dow Jones rallied towards the end of the year, it closed 34 percent down a bone-crus.h.i.+ng decline in the apparent wealth of America's stockholders (see Figure 13.10).

13.9 The change in debt collapses as the Great Recession begins 13.10 The Dow Jones nosedives The long bubble in the housing market which neocla.s.sical economists like Ben Bernanke had strenuously denied was a bubble burst under the weight of sheer fraud involved in subprime lending, well before the debt bubble propelling it started to slow.9 It continued its decline relentlessly in 2008/09, with house prices falling another 19 percent (in real terms) on top of the 10 percent decline from their peak in March 2006 see Figure 13.12.

Unemployment rose from 4.4 percent at the beginning of 2007 to 5.5 percent at its end, and then to 7.6 percent as 2009 began. Here the hand of debt was clearly visible, for the simple reason that, since the change in debt is a major component of aggregate demand, and aggregate demand determines employment, unemployment rises if the rate of change of debt falls (and vice versa). As the level of debt has risen relative to GDP, the ebb and flow of unemployment has fallen more and more under the sway of changes in the level of private debt.

13.11 The correlation of debt-financed demand and unemployment 13.12 The housing bubble bursts 13.13 The Credit Impulse and change in employment The dominance of debt has been obvious, not only in the collapse into the Great Recession, but even in the apparent recovery from it in late 2010 and early 2011 (a recovery that I believe will prove temporary, and which is also exaggerated by unreliable government statistics). Here an apparent paradox emerges: because aggregate demand is the sum of GDP plus the change in debt, the rate of change of aggregate demand can be boosted by a slowdown in the rate at which debt is falling.

The logic here is a simple extrapolation from the observation that the level of aggregate demand is the sum of GDP10 plus the change in debt: given this, the change in aggregate demand is equal to the change in GDP plus the acceleration of debt. Therefore the factor that determines debt's impact upon the rate of economic growth and hence the change in the rate of unemployment is not the rate of change of debt, but the rate of change of its rate of change.

Biggs, Mayer and Pick, who first made this observation, noted that it had a seemingly counter-intuitive outcome that the economy can receive a boost from credit, even if the aggregate level of debt is falling, so long as the rate of that fall decreases: 'the flow of credit and GDP can increase even while the stock of credit is falling' (Biggs, Mayer et al. 2010: 5). They measured the impact of the acceleration of credit on changes in aggregate demand using the ratio of the acceleration of debt to GDP (which they termed 'the Credit Impulse'; ibid.: 3), and this measure clearly ill.u.s.trated their apparently bizarre conclusion that the slight recovery in late 2010 was driven in large measure by a slowdown in the rate of deceleration of credit see Figure 13.13.11 There are thus three factors that need to be considered to understand the impact of debt on a capitalist economy: the level of debt, the rate of change of debt, and its rate of acceleration all measured with respect to the level of GDP.

Box 13.1 Definitions of unemployment The official definition of unemployment has been reworked numerous times, in ways that reduce the recorded number, so much so that the published levels drastically understate the actual level. The official OECD definition (see stats.oecd.org/glossary/detail.asp?ID=2791) requires that those recorded as unemployed must be both available for work and actively looking for work in the reference period, which excludes those who have become discouraged by the sheer unavailability of employment opportunities during a major recession, but many OECD countries have further tailored the definition to reduce the recorded numbers.

The Australian government's definition is typical here: in addition to the OECD requirements, it also records as employed people who 'worked for one hour for pay, profit, commission or payment in kind in a job or business, or on a farm; or worked for one hour or more without pay in a family business or on a farm' (McLennan 1996: 47). To regard someone who has worked only one hour in a week as employed is simply absurd at least fifteen hours of work at the minimum wage are needed to be paid even the equivalent of unemployment benefits.

Similar distortions apply in other countries. The USA, for example, ceases counting someone as unemployed if they have been out of work for more than a year a change in definition introduced in 1994 (see en.wikipedia.org/wiki/Unemployment#

United_States_Bureau_of_Labor_Statistics and en.wikipedia.org/wiki/Current_Population

_Survey#Employment_cla.s.sification for more details). Abuses of statistics like this have prompted private citizens to record what official statistics ignore. The opinion-polling organization Roy Morgan Research (www.roymorgan.com.au/) now publishes its own survey of Australian unemployment, which it puts at 7.9 percent versus the recorded figure of 5.5 percent (the not-seasonally-adjusted figure as of January 2011).

Shadowstats (www.shadowstats.com/alternate_data/unemploymentcharts) maintains an alternative measure for the USA that includes long-term discouraged workers. This is now more than twice as high as the official US measure: at the time of writing (February 2011), the official U-3 measure was 9.0 percent, while the Shadowstats measure was 22.2 percent.

This, plus changes in the structure of employment, make comparisons with past economic crises like the Great Depression very difficult. John Williams, the founder of Shadowstats, estimates that his measure of unemployment would have shown that 3435 percent of the workforce was unemployed during the Great Depression versus the 25 percent actually recorded back then, since the proportion of the population working on farms was much higher in the 1930s than now (27 percent then versus 2 percent now). The workers who were underemployed on farms but nonetheless fed reduced the numbers officially recorded as unemployed back then.

Given these problems, I regard the US's U-6 measure of unemployment today which includes those who have been unemployed for two years or less as more comparable to the Great Depression figures than its U-3 measure, which omits those who have been unemployed for a year or more. On that basis, one in six Americans are out of work today, versus the peak rate of one in four during the Great Depression. The current crisis, though it is called the Great Recession, is therefore really a depression too.

The first factor indicates the aggregate burden that debt imposes upon society. Since the level of debt is a stock, while the level of GDP is a flow (of income per year), the ratio tells us how many years of income it would take to reduce debt to zero. Of course, a target of zero debt is neither feasible nor desirable as explained earlier, some debt is necessary to support entrepreneurial innovation. But the ratio indicates how debt-enc.u.mbered an economy has become, and the larger it is, the longer it will take to get back to any desired lower level.

It also provides the best measure of the burden the financial sector imposes upon the economy, since the net cost of the financial sector is the level of debt (multiplied by the inflation-adjusted gap between the rate of interest on loans and that on deposits a gap that has been relatively constant, though the nominal and real rates of interest themselves have been very volatile).

The second factor indicates how much aggregate demand is being generated by rising debt or reduced by falling debt. When the economy is growing, so too will credit, and again this is not a bad thing when that debt finances investment. The danger arises when the rate of growth of debt becomes a substantial determinant of overall demand as it has in the Ponzi economy the USA has become. A large debt-financed contribution to aggregate demand will almost certainly have a large component of Ponzi finance behind it, and such an increase necessarily requires a decline in debt-financed spending in the near future, which will usher in a recession.

The third factor is the best leading indicator of whether employment and the economy are likely to grow in the near future. The Credit Impulse leads both changes in GDP and changes in employment, with the lead (in the USA) being about two months to employment and four months to GDP.

13.14 Correlation of credit impulse and change in employment and GDP The Credit Impulse is also the key financial source of capitalism's inherently cyclical nature. To maintain a stable rate of employment, the rate of growth of aggregate demand has to equal the rate of growth of employment and labor productivity, which are both relatively stable. But since the rate of growth of aggregate demand depends on the rate of growth of GDP and the acceleration of debt, a stable rate of growth of aggregate demand requires a constant acceleration of debt.

The only level at which this is possible is zero. Just as maintaining a constant positive rate of acceleration while driving a car is impossible since otherwise the car would ultimately be travelling faster than the speed of light a constant positive rate of acceleration of debt can't be maintained, because this would mean that debt would ultimately be infinitely larger than GDP. Since in the real world it is impossible for the acceleration of debt to always be zero, the economy will therefore necessarily have cycles driven by the expansion and contraction of credit.

These three factors the level of debt, its rate of change, and its acceleration interact in complex ways that are best explained by an a.n.a.logy to driving in which the debt-to-GDP ratio is like the distance back to your starting point, its rate of change relative to GDP is like the speed of the car, and the Credit Impulse is like the car's acceleration or deceleration.

A low ratio of debt to GDP is like having taken a short drive say, from Los Angeles to Phoenix (a distance of 370 miles). It's easy to get back to LA at any time, and the return journey is not something one has to plan all that much for. A high ratio is like a drive from LA to New York: it's a huge distance (2,800 miles), and the drive back which corresponds to reducing the debt to GDP ratio will take a long time.