Debtwatch No 34: The Confidence Trick

“And, at this point, confidence is what it is all about… The first thing is to maintain some confidence in ourselves and the prospects for our country over time… Unfortunately, there is no lever marked ‘confidence’  that policy-makers can take hold of. Our task is very much one of seeking to behave, across the board, in ways that will foster, rather than erode, confidence.  It is such confidence that, more than anything else, will help to drive us along the road to recovery.” (Glenn Stevens, April 21st 2009)“I fancy that over-confidence seldom does any great harm except when, as, and if, it beguiles its victims into debt.” (Irving Fisher, 1933)

In his recent speech “The Road To Recovery“, Australia’s Reserve Bank Governor Glenn Stevens used “the C word” 17 times–versus, for example, 15 uses of the “R” word (”recession”). The message was clearly that, if only we can all be confident, then the other “R” word (”recovery”–which received ten mentions) will surely occur.

Another prominent economist who had the same attitude at the outbreak of a financial crisis was Irving Fisher. Speaking to a bankers conference just two days before the Great Crash of 1929, Fisher argued that market downturns were caused by a “lunatic fringe”. Once they had exited, the bull market of the preceding years would resume:

“There is a certain lunatic fringe in the stock market, and there always will be whenever there is any successful bear movement going on… they will put the stocks up above what they should be and, when frightened, … will immediately want to sell out… when it is finally rid of the lunatic fringe, the stock market will never go back to 50 per cent of its present level…

We shall not see very much further, if any, recession in the stock market, but rather … a resumption of the bull market, not as rapidly as it has been in the past, but still a bull rather than a bear movement.”  (Fisher 1929)

Fisher’s confidence led him to hang on to his margin-financed stocks (worth over $100 million in 2000-dollar terms). Despite his confidence, the stock market continued its plunge from its peak of 31.3 in July 1929 to the nadir of 4.77 in May of 1932, while unemployment rose from zero to 25 percent. Fisher was wiped out financially, and left to ponder how he could have got the behaviour of the market, and the economy, so badly wrong.


Three years later, he reached the conclusion that he had been misled by two core elements of the neoclassical theory he had helped build: the beliefs that the economy was always in equilibrium, and that the debt commitments borrowers had entered into to purchase financial assets were based on correct forecasts of future economic prospects.

On equilibrium, he reasoned, even if it were true that the economy tended towards equilibrium, random events alone would ensure that all economic variables were either above or below their equilibrium levels. Therefore economic theory had to be about disequilibrium rather than equilibrium:

“Theoretically there may be— in fact, at most times there must be—  over- or under-production, over- or under-consumption, over- or under spending, over- or under-saving, over- or under-investment, and over or under everything else. It is as absurd to assume that, for any long period of time, the variables in the economic organization, or any part of them, will “ stay put,”  in perfect equilibrium, as to assume that the Atlantic Ocean can ever be without a wave.” (Fisher 1933)

This realisation in turn put paid to any notion that today’s debt commitments were based on an accurate prediction of tomorrow’s economic outcomes. Instead, he identified over-indebtedness as one of the two key causes of Great Depression:

“two dominant factors [are …] over-indebtedness to start with and deflation following soon after… these two economic maladies, the debt disease and the price-level disease, are, in the great booms and depressions, more important causes than all others put together.

Thus over-investment and over-speculation are often important; but they would have far less serious results were they not conducted with borrowed money. That is, over-indebtedness may lend importance to over-investment or to over-speculation.

The same is true as to over-confidence. I fancy that over-confidence seldom does any great harm except when, as, and if, it beguiles its victims into debt.” (Fisher 1933)

One would hope that economic theory had learnt from the Great Depression, and in particular from Fisher’s insights. Unfortunately, economics was eager to unlearn these lessons, because the very phenomenon of a Depression was anathema to a profession that had always sought to eulogise the market economy, rather than to understand it. Equilibrium came back again in the guise of the “Neoclassical-Keynesian synthesis” in the 1950s. By the 1990s, all vestiges of Keynes had been thrown away–and nothing of Fisher had been even assimilated in the first place (skerricks of his thought are percolating through now though: see The Economist for a pretty good overview of Fisher).

Today, macroeconomic models like TRYM (the TReasurY Macroeconomic model that is used to prepare the Australian Federal Budget) presume that the economy tends towards a “long run equilibrium”. The apparent dynamics such models display are simply the convergence of the model from an initial starting point to the assumed long run equilibrium.

For example, the figure below shows the TRYM model’s predictions for unemployment from March 1995 till March 2010 (Figure 10: Dynamic Adjustment towards Steady State – Unemployment; Modelling Section, Macroeconomic Analysis Branch, Commonwealth Treasury, The Macroeconomics Of The Trym Model Of The Australian Economy,  Commonwealth of Australia 1996). The model “predicted” that unemployment would fall from around 9 percent in 1995 to just below 7 percent in 2010, simply on the basis that unemployment was assumed to converge to an a long run equilibrium rate of 7 percent over time (the actual level fell well below this, and the  assumed equilibrium unemployment rate–the “NAIRU”–was therefore later reduced to 5.25 percent).


Virtually everyone knows Keynes’s quip that “in the long run we are all dead”. Yet very few realise that Keynes’s target was precisely this approach to economic modelling–of assuming that the economy would simply tend to return to equilibrium after any disturbance:

“ But this long run is a misleading guide to current affairs. In the long run we are all dead. Economists set themselves too easy, too useless a task if in tempestuous seasons they can only tell us that when the storm is long past the ocean is flat again.”   (Keynes, 1923)

Hobbled by this naive belief in equilibrium, the economics profession was as unprepared for today’s crisis as it had been for the Great Depression. Now that the crisis is well and truly with us, all conventional “neoclassical” economists can offer is the hope that the crisis can be overcome by a good, strong dose of confidence.

From Fisher’s point of view, such a belief is futile. In an economy with an excessive level of debt and low inflation, he argued that confidence was irrelevant–and in fact dangerously misleading, as he knew from painful personal experience. Given over-indebtedness and low levels of inflation, a “chain reaction” would occur in which:

“(1) Debt liquidation leads to distress selling and to

(2) Contraction of deposit currency, as bank loans are paid off, and to a slowing down of velocity of circulation. This contraction of deposits and of their velocity, precipitated by distress selling, causes

(3) A fall in the level of prices, in other words, a swelling of the dollar. Assuming, as above stated, that this fall of prices is not interfered with by reflation or otherwise, there must be

(4) A still greater fall in the net worths of business, precipitating bankruptcies and

(5) A like fall in profits, which in a “ capitalistic,”  that is, a private-profit society, leads the concerns which are running at a loss to make

(6) A reduction in output, in trade and in employment of labor. These losses, bankruptcies, and unemployment, lead to

(7) Pessimism and loss of confidence, which in turn lead to

(8) Hoardinq and slowing down still more the velocity of circulation.The above eight changes cause

(9) Complicated disturbances in the rates of interest, in particular, a fall in the nominal, or money, rates and a rise in the real, or commodity, rates of interest.” (Fisher 1933; The Debt Deflation Theory of Great Depressions)

One key phenomenon that Fisher emphasised was that deflation could make the debt burden worse even as borrowers reduced their nominal debt levels–something I have termed “Fisher’s Paradox”. In Fisher’s words:

“Each dollar of debt still unpaid becomes a bigger dollar, and if the over-indebtedness with which we started was great enough, the liquidation of debts cannot keep up with the fall of prices which it causes.

In that case, the liquidation defeats itself. While it diminishes the number of dollars owed, it may not do so as fast as it increases the value of each dollar owed.

Then, the very effort of individuals to lessen their burden of debts increases it, because of the mass effect of the stampede to liquidate in swelling each dollar owed. Then we have the great paradox which, I submit, is the chief secret of most, if not all, great depressions:

The more the debtors pay, the more they owe. The more the economic boat tips, the more it tends to tip. It is not tending to right itself, but is capsizing.”

This is a “disequilibrium” phenomenon par excellence, because not only does it occur out of equilibrium, it actually drives the system further from equilibium. And it is indeed what happened during the Great Depression: America’s debt to GDP ratio rose even as nominal debt levels were reduced. The debt ratio rose from 175 at the end of 1929 to 235 percent in 1932, even as nominal private debt fell from US$163 billion to US$134 billion.


Even though the public’s initial attempt to reduce its debt burden was foiled, the reduction in debt nonetheless did have an impact: it drove the economy into Depression. In the credit-driven real world in which we live, aggregate demand is the sum of GDP plus the change in debt. The public’s attempt to reduce debt meant that the reductions in debt substantially reduced demand, and this deleveraging was the unstoppable force that made the Great Depression “great”.

As the next chart shows, during the Roaring Twenties, the annual increase in debt was responsible for up to ten percent of aggregate demand. But when the Great Crash brought this period of leveraged speculation to an end, the deleveraging that Fisher described meant that the change in debt started to reduce from demand–and at its peak, the reduction in debt in 1932 reduced aggregate demand by 25 percent.


As is obvious, unemployment skyrocketed as aggregate demand collapsed. When debt reaches the sky high levels it did before the Great Depression, deleveraging becomes the dominant force determining the level of unemployment–but obviously there is a lag. Unemployment is the classic “lagging indicator”, because firms take time to respond to a drop in demand, firstly by ceasing to hire new workers and then by sacking existing ones.

When working with annual data at the time of the Great Depression, this lag appears to be about one and a half years. Applying that lag to the period from mid-1929 till mid-1938 (when Government spending and armaments production for the looming war in Europe started to boost demand and caused unemployment to fall), the correlation between debt’s contribution to demand and unemployment was -0.85. The change in debt’s contribution to demand thus explains 85 percent of the unemployment experience of the Great Depression.


This is not good news for us today, for three reasons. Firstly, debt levels today are far higher than they were prior to the Great Depression–the force of deleveraging is thus likely to be greater now than it was in the 1930s. Secondly, given this higher level of debt, the correlation between the debt-financed proportion of aggregate demand and unemployment is even stronger now than it was during the Great Depression. Thirdly, given the greater dependence on debt today than ever before, and the social changes that have gone with the Ponzification of Capitalism, the lag between a fall in the debt-financed component of demand and a rise in unemployment has dropped to just two months.


The change in debt is therefore the best–and most ominous–predictor of future unemployment levels. Though well down from the peak level of being responsible for 25% of aggregate demand, private debt is still generating 10 percent of demand in the USA. Yet even with still positive debt-financed demand, unemployment has risen to 8.7 percent. If deleveraging results in debt reducing aggregate demand by 25 percent as it did in the Great Depression, then unemployment is going to go much, much higher.

The same analysis applies to Australia. Since the crisis has yet to hit Australia as strongly as it has the USA or Europe, the belief that “we are different”–which I call “Kangaroo Economics” in honour of our national fauna–is still prevalent here. So too is the belief that, if we do suffer a recession, it will be due to external forces rather than to our own economic circumstances.

The data begs to differ. Though our aggregate debt level didn’t reach Yankee heights–our peak debt to GDP level was about 165%, versus 290% in the USA before deflation started–our rate of growth of debt was much higher, so that at its peak the growth in debt was responsible for 22% of aggregate demand. Now that debt is starting to fall, unemployment is starting to rise. There is every reason to expect deleveraging in Australia to drive unemployment well into double digits.


So confidence is not “all it is about”: confidence played its role over the last thirty years as it “beguiled its victims into debt”, in Fisher’s evocative phrase. We don’t need more of it now, so much as less of it back then–but of course, we can’t amend history.

The victims of past overconfidence include Central Bankers, whose rescues of the financial system simply encouraged it to search out a new group of potential borrowers to replace those who had already been debt-saturated. They were victims of debt, as much as were the borrowers, because the naive theory of economics they followed ignored the role of debt completely. They therefore couldn’t see the process that was leading to crisis, even as their interventions egged that process on to heights that it could never have reached without them.

Had Greenspan and his equivalents around the world not intervened in 1987, it is quite possible that we would have experienced a mild Depression back then–mild because debt was only equivalent to 1929 levels then, because a larger Government sector than in the 1920s would have counterbalanced the private sector downturn, and because higher inflation in the late 80s would have helped reduced the real burden of debt.

Now we are sitting on the precipice of a mountain of debt twice as high as in the Great Depression, with low inflation turning into deflation as Fisher warned, and with Central Bankers who do not have a clue why the economy has suddenly gone from “the Great Moderation” to “the Greatest Crisis Since the Great Depression”.

Over-confidence in the face of rising debt did beguile us during the long boom. Confidence in the face of deleveraging will not save us during the coming Depression.


Comments on the Australian Data

Debt levels in Australia are very close to falling in nominal terms, and in fact only mortgage debt is still rising: both business and personal debt (other than mortgages) have fallen in the last few months. It is conceivable that, were it not for the “First Home Buyers Boost”, mortgage debt as well would be falling now too (the scheme is more aptly described as the “First Home Vendors Boost”, since prices at the low end of the market have been driven up by far more than the $7,000 increase in the grant).

As a result, the debt to GDP ratio has fallen for the last four months–though this is to some extent masked by Australia’s practice of summing the previous four quarters of GDP data to derive annual GDP, versus the American practice of simply multiplying the current quarter’s GDP figure by 4. Using the Australian approach, our debt to GDP ratio is now 160%; using the American, it is 162%, since GDP fell by 0.5% in the previous quarter.

Whichever way you cut it, deleveraging is now well and truly underway, and unemployment will therefore rise dramatically in the next few months. Most neoclassical economists are predicting 7.5% unemployment by mid-2010; I expect it will have entered double figures by early in 2010.

Table One

IMG0019_16465156.PNGTable Two


Originally published at Steve Keen’s Oz Debtwatch blog and reproduced here with the author’s permission.

2 Responses to "Debtwatch No 34: The Confidence Trick"

  1. Guest   May 4, 2009 at 6:27 pm

    If you knew the amount of bad debt Wells has on its books (and this is just one bank), you would triple the rate of unemployment you are predicting.You really have no idea how bad it really is. Expect a collapse by late August. People think the Administration is intervening heavily. The reality is that it has given up hope.Too bad people don’t really know what is going on inside.John Ryskamp

  2. Ike   May 11, 2009 at 10:30 am

    This is a fascinating aspect of the excess leverage in our system and what unwinding it can lead to; deflation could be with us for awhile and is a scary possibility.