Eco­nom­ics in the Age of Delever­ag­ing

Flattr this!

Click here for this post in PDF: Debt­watch mem­bers; CfESI mem­bers
Click here for the data in this post: Debt­watch mem­bers; CfESI mem­bers

The “Global Finan­cial Cri­sis”, which began in late 2007, marked a turn­ing point in the nature of mar­ket economies. Their per­for­mance from at least the mid-1960s had been under-writ­ten by a faster growth of pri­vate debt than of GDP: this was the “Age of Lever­age”. In late 2007, the growth rate of pri­vate debt fell, and since then we have been in the Age of Delever­ag­ing.

The sta­tis­tics of the Ages are stark enough: pri­vate debt rose six­fold com­pared to GDP in Amer­ica from 1945, and six­fold in Aus­tralia from 1965. Pre-1988 fig­ures aren’t avail­able for UK debt, but clearly it has exploded since 1988. All three ratios peaked after 2007, and have since been falling. The fall in the US ratio is clearly unprece­dented in the post-WWII period: only the Great Depres­sion com­pares.

Fig­ure 1

The change in debt data is just as stark—and it points out one sub­stan­tive dif­fer­ence between Aus­tralia on the one hand and the USA & UK and most of the West­ern OECD on the other. Aus­tralian pri­vate debt is still ris­ing (though more slowly than nom­i­nal GDP), whereas US pri­vate debt has been falling in absolute terms, and the UK has fluc­tu­ated between ris­ing and falling debt.

Fig­ure 2

Non-econ­o­mists might expect pro­fes­sional econ­o­mists to pay great heed to these indicators—after all, surely pri­vate debt affects the econ­omy? How­ever, the dom­i­nant approach to economics—known as “Neo­clas­si­cal Eco­nom­ics” —ignores them com­pletely, on the a pri­ori grounds that the aggre­gate level of pri­vate debt doesn’t mat­ter: only its dis­tri­b­u­tion can have macro­eco­nomic impacts.

The argu­ment is that a rise in debt merely indi­cates a trans­fer of spend­ing power from a saver to a bor­rower. The debtor’s spend­ing power rises, but saver’s spend­ing power also falls, so in the aggre­gate there will only be a macro­eco­nomic effect if there is a very large dif­fer­ence in behav­iour between the saver and bor­rower. There­fore only the dis­tri­b­u­tion of debt mat­ters, not its level or rate of change.

US Fed­eral Reserve Chair­man Ben Bernanke pro­vided pre­cisely this ratio­nale to explain why neo­clas­si­cal econ­o­mists ignored Irv­ing Fisher’s “debt-defla­tion” expla­na­tion of the Great Depres­sion (Fisher 1933), and he also asserted that the dif­fer­ences in behav­iour between saver and bor­rower could not be large enough to explain the Great Depres­sion:

Fisher’s idea was less influ­en­tial in aca­d­e­mic cir­cles, though, because of the coun­ter­ar­gu­ment that debt-defla­tion rep­re­sented no more than a redis­tri­b­u­tion from one group (debtors) to another (cred­i­tors). Absent implau­si­bly large dif­fer­ences in mar­ginal spend­ing propen­si­ties among the groups, it was sug­gested, pure redis­tri­b­u­tions should have no sig­nif­i­cant macro-eco­nomic effects… (Bernanke 2000, p. 24)

Sim­i­larly, Nobel Prize win­ner Paul Krug­man argued recently that the aggre­gate level of pri­vate debt was not a fac­tor in the GFC: only its dis­tri­b­u­tion could be. He there­fore devel­oped a model in which the dis­tri­b­u­tion of debt, rather than its level, was the causal fac­tor:

Ignor­ing the for­eign com­po­nent, or look­ing at the world as a whole, the over­all level of debt makes no dif­fer­ence to aggre­gate net worth — one person’s lia­bil­ity is another person’s asset…

In what fol­lows, we begin by set­ting out a flex­i­ble-price endow­ment model in which “impa­tient” agents bor­row from “patient” agents, but are sub­ject to a debt limit. If this debt limit is, for some rea­son, sud­denly reduced, the impa­tient agents are forced to cut spend­ing… (Krug­man and Eggerts­son 2010, p. 3)

He reasserted this analy­sis in a recent blog, argu­ing that the level of debt doesn’t mat­ter, because most debt is “money we owe to our­selves”:

Peo­ple think of debt’s role in the econ­omy as if it were the same as what debt means for an indi­vid­ual: there’s a lot of money you have to pay to some­one else. But that’s all wrong; the debt we cre­ate is basi­cally money we owe to our­selves, and the bur­den it imposes does not involve a real trans­fer of resources.

That’s not to say that high debt can’t cause prob­lems — it cer­tainly can. But these are prob­lems of dis­tri­b­u­tion and incen­tives, not the bur­den of debt as is com­monly under­stood. (Krug­man 2011)

Bizarre as it may sound, these argu­ments by lead­ing econ­o­mists ignore decades of empir­i­cal research into and prac­ti­cal knowl­edge on bank­ing, which have estab­lished that their fun­da­men­tal premise is false: a new debt is not a trans­fer from one bank customer’s account to another’s—which is effec­tively what Krug­man mod­els and Bernanke assumes above—but a simul­ta­ne­ous cre­ation of both a deposit and a debt by the bank. A bank loan thus gives a bor­rower addi­tional spend­ing power with­out forc­ing savers to reduce their spend­ing power to com­pen­sate. As Joseph Schum­peter put it dur­ing the Great Depres­sion:

It is always a ques­tion, not of trans­form­ing pur­chas­ing power which already exists in someone’s pos­ses­sion, but of the cre­ation of new pur­chas­ing power out of noth­ing…’ (Schum­peter 1934, p. 73; see Keen 2011 , pp. 154–157, for more detail on this issue)

From this empir­i­cally con­firmed per­spec­tive (Holmes 1969; Moore 1979; Kyd­land and Prescott 1990; Car­pen­ter and Demi­ralp 2010), the change in debt there­fore does have seri­ous macro­eco­nomic con­se­quences, since an increase in debt adds to aggre­gate demand—and it is the pri­mary means by which both invest­ment (Fama and French 2002) and spec­u­la­tion (Min­sky 1982, pp. 28–30) are funded.

In a credit-based econ­omy, aggre­gate demand is there­fore the sum of income plus the change in debt, with the change in debt spend­ing new money into exis­tence in the econ­omy. This is then spent not only goods and ser­vices, but on finan­cial assets as well—shares and prop­erty. Changes in the level of debt there­fore have direct and poten­tially enor­mous impacts on the macro­econ­omy and asset mar­kets, as the GFC—which was pre­dicted only by a hand­ful of credit-aware econ­o­mists (Beze­mer 2009)—made abun­dantly clear.

If the change in debt is roughly equiv­a­lent to the growth in income—as applied in Aus­tralia from 1945 to 1965, when the pri­vate debt to GDP ratio fluc­tu­ated around 25 per cent (see Fig­ure 1)—then noth­ing is amiss: the increase in debt mainly finances invest­ment, invest­ment causes incomes to grow, and the econ­omy moves for­ward in a vir­tu­ous feed­back cycle. But when debt rises faster than income, and finances not just invest­ment but also spec­u­la­tion on asset prices, the vir­tu­ous cycle gives way to a vicious pos­i­tive feed­back process: asset prices rise when debt rises faster than income, and this encour­ages more bor­row­ing still.

The result is a super­fi­cial eco­nomic boom dri­ven by a debt-financed bub­ble in asset prices. To sus­tain a rise in asset prices rel­a­tive to con­sumer prices, debt has to grow more rapidly than income—in other words, if asset prices are to rise faster than con­sumer prices, then rather than merely ris­ing, debt has to accel­er­ate. This in turn guar­an­tees that the asset price bub­ble will burst at some point, because debt can’t accel­er­ate for­ever. When debt growth slows, a boom can turn into a slump even if the rate of growth of GDP remains con­stant.

This process is eas­ily illus­trated in a numer­i­cal exam­ple. Con­sider an econ­omy with a GDP of $1 tril­lion that is grow­ing at 10% per annum, with real growth of 5% and infla­tion of 5%, and in which pri­vate debt is $1.25 tril­lion and grow­ing at 20% p.a. Total spend­ing on both goods & ser­vices and finan­cial assets is there­fore $1.25 tril­lion: $1 tril­lion is financed by income, and $250 bil­lion is financed by the 20% increase in debt.

In the fol­low­ing year, if the growth of debt sim­ply slows down to the same rate at which nom­i­nal GDP is grow­ing (with­out affect­ing the rate of eco­nomic growth), then the growth in debt will be $150 bil­lion (10% of the $1.5 tril­lion level reached at the end of the pre­vi­ous year). Total spend­ing will there­fore be exactly the same as the year before: $1.25 tril­lion, con­sist­ing of $1.1 tril­lion in GDP plus a $150 bil­lion growth in debt. How­ever, since infla­tion is run­ning at 5%, this amounts to a 5% fall in the real level of eco­nomic activity—which would be spread across both com­mod­ity and asset mar­kets.

If instead the growth of debt stopped, then total spend­ing the next year will be $1.1 tril­lion, a 15% fall from the level of the pre­vi­ous year in nom­i­nal terms, and 20% in real terms. This would cause a mas­sive slump in demand for goods & ser­vices, assets, or both, even with­out a slow­down in the rate of growth of GDP.

This hypo­thet­i­cal exam­ple is not far removed from the actual expe­ri­ence of the GFC. As the US expe­ri­ence illus­trates most clearly, the switch from ris­ing to falling pri­vate debt ush­ered in the biggest eco­nomic down­turn since the Great Depres­sion, a pro­longed period of high unem­ploy­ment, and sharp falls in asset markets—all of which are plot­ted in Fig­ure 3.

Fig­ure 3

This is why the shift from the Age of Lever­age to the Age of Delever­ag­ing was so dra­matic, and yet so unfore­seen by con­ven­tional econ­o­mists: it was caused by a huge reduc­tion in aggre­gate demand from a fac­tor they ignore. This debt-induced reduc­tion in aggre­gate demand will per­sist as long as pri­vate debt lev­els are falling—as they still are in the USA, though at a much reduced rate from the peak rate of fall in early 2010.

Until pri­vate debt lev­els are sub­stan­tially reduced, the econ­omy will always tend towards what Richard Koo called a “bal­ance sheet reces­sion” (Koo 2009), where the desire of the pri­vate sec­tor to reduce its lever­age will sup­press aggre­gate demand, caus­ing both reces­sions and falling asset prices. The West­ern OECD is thus “turn­ing Japanese”—replicating the cri­sis that led to Japan’s “Lost Decade”, which is now two decades old.

Koo argues that whether the world expe­ri­ences the rel­a­tively minor decline in Japan­ese eco­nomic per­for­mance or achieves some­thing much worse depends in part on whether the world mim­ics Japan’s pol­icy response or not. But whereas con­ven­tional wis­dom argues that Japan has failed by run­ning huge gov­ern­ment deficits, Koo argues that with­out these deficits, Japan­ese GDP would have fallen far more.

His rea­son­ing is that, just as pri­vate sec­tor bor­row­ing spends addi­tional money into exis­tence, so too does a gov­ern­ment deficit. But if the pri­vate sec­tor is deleveraging—as it has done in Japan since 1991 and is now doing in the USA—then the change in pri­vate debt is actu­ally sub­tract­ing from demand. Japan’s pub­lic sec­tor deficits there­fore atten­u­ated the decline in aggre­gate demand:

Although this fis­cal action increased gov­ern­ment debt by … 92 per­cent of GDP dur­ing the 1990–2005 period, the amount of GDP pre­served by fis­cal action com­pared with a depres­sion sce­nario was far greater. For exam­ple, if we assume, rather opti­misti­cally, that with­out gov­ern­ment action Japan­ese GDP would have returned to the pre-bub­ble level of 1985, the dif­fer­ence between this hypo­thet­i­cal GDP and actual GDP would be over 2,000 tril­lion yen for the 15-year period. In other words, Japan spent 460 tril­lion yen to buy 2,000 tril­lion yen of GDP, mak­ing it a tremen­dous bar­gain. And because the pri­vate sec­tor was delever­ag­ing, the government’s fis­cal actions did not lead to crowd­ing out, infla­tion, or sky­rock­et­ing inter­est rates. (Koo 2011, p. 23)

On the other hand, Koo cau­tions that if the gov­ern­ment attempts to run a sur­plus while the pri­vate sec­tor is delever­ag­ing, there will be two fac­tors reduc­ing eco­nomic activ­ity at the same time. He there­fore argues that deficits are sen­si­ble when the pri­vate sec­tor is delever­ag­ing, while attempt­ing to run sur­pluses will make a bad sit­u­a­tion worse:

Although shun­ning fis­cal profli­gacy is the right approach when the pri­vate sec­tor is healthy and is max­i­miz­ing prof­its, noth­ing is worse than fis­cal con­sol­i­da­tion when a sick pri­vate sec­tor is min­i­miz­ing debt. (Koo 2011, p. 27)

How­ever, fis­cal con­sol­i­da­tion is the pol­icy pre­scrip­tion that is being applied in the Euro zone and the UK, and sup­ported by politi­cians in both Aus­tralia and the USA. The likely out­come of pub­lic aus­ter­ity is thus a fur­ther decrease in the growth rate of coun­tries prac­tic­ing it. Given that most of the West­ern OECD is already under severe eco­nomic stress, the pain that aus­ter­ity inflicts upon already stressed soci­eties is likely to mean dras­tic polit­i­cal change.

The most obvi­ous loca­tion for polit­i­cal tur­moil is Europe, where the Maas­tricht Treaty’s rules force coun­tries to attempt to restrain fis­cal deficits to 3% of GDP. This was always a bad idea, pred­i­cated on the belief that severe eco­nomic crises could not occur. Though the treaty was applauded by neo­clas­si­cal econ­o­mists, I was far from the only non-neo­clas­si­cal econ­o­mist to observe that this treaty could lead to the breakup of Europe when a reces­sion hit, since “Europe’s gov­ern­ments may be com­pelled to impose aus­ter­ity upon economies which will be in des­per­ate need of a stim­u­lus” (Keen 2001, pp. 212–13; see also Keen 2011, pp. 2–3).

Though con­ti­nen­tal Europe is the most obvi­ous loca­tion for eco­nom­i­cally inspired polit­i­cal insta­bil­ity in 2012, another dark horse may also be the UK. As Fig­ure 1 shows, its pri­vate debt level is stag­ger­ingly high—one and a half times the peak level of the USA’s—and yet it did not suf­fer as severe a down­turn as the USA when the cri­sis began because, as Fig­ure 2 indi­cates, it did not fall as deeply into delever­ag­ing. The max­i­mum decline in aggre­gate demand caused by falling pri­vate debt in the UK was only 6 per­cent of GDP, ver­sus 20 per­cent in the USA.

How­ever the rate of delever­ag­ing in the UK has again hit this level, while the USA has recov­ered from the worst of the ini­tial down­turn and is delever­ag­ing at a rate of only 3% of GDP. Gov­ern­ments in both the USA and the UK are favour­ing aus­ter­ity poli­cies, but the UK is already impos­ing them—with thus far neg­a­tive results—and is far more likely to be able to main­tain them than the polit­i­cally ham­strung USA. This means that pri­vate sec­tor delever­ag­ing and pub­lic sec­tor aus­ter­ity may coin­cide in the UK in 2012, which from a “bal­ance sheet reces­sion” per­spec­tive indi­cates that the UK could fall into reces­sion from an already depressed level of eco­nomic activ­ity. This is espe­cially likely if the rate of pri­vate sec­tor delever­ag­ing accel­er­ates.

What could the future hold for Aus­tralia? To date, sangfroid has dom­i­nated Aus­tralian atti­tudes towards the GFC—based partly on our avoid­ance of a deep down­turn in 2008, which was unique amongst OECD nations, and partly on our lucky depen­dence on China. At the begin­ning of 2011, the RBA expected to be rais­ing rates to restrain infla­tion in a boom­ing econ­omy, while Trea­sury expected unem­ploy­ment to fall towards full employ­ment lev­els. Eco­nomic poli­cies pro­posed by the major polit­i­cal par­ties were based on expec­ta­tions of man­ag­ing a boom, and the only debate was over how quickly the Bud­get should return to sur­plus.

Fig­ure 4: Trea­sury fore­casts (& pro­jec­tions of a return to equi­lib­rium) in the 2011-12 bud­get

Unfor­tu­nately, recent eco­nomic data hasn’t fol­lowed the sangfroid script. Inflation—as mea­sured by the RBA’s pre­ferred indi­ca­tors, the weighted and trimmed means—has fallen rather than risen, while unem­ploy­ment has risen from a low of 4.9% to 5.3% (ver­sus expec­ta­tions of 4.75% in June 2012) and appears likely to trend up rather than down in com­ing months.

If this does hap­pen, it will not be an indi­ca­tion that Gov­ern­ment deficits are the problem—or even that they have failed to stim­u­late the economy—but a cau­tion that Aus­tralia is not so dif­fer­ent after all. Though it was delayed by pol­icy and the min­ing boom, Aus­tralia is now on the cusp of a bal­ance sheet reces­sion too.

To see this, we need to take a closer look at the Aus­tralian pri­vate debt level. Fig­ure 5 con­sid­ers both the level (com­pared to GDP) and rate of change of Aus­tralian pri­vate debt since 2000.

Fig­ure 5


After ini­tially falling in 2008, Australia’s pri­vate debt to GDP ratio actu­ally rose from mid-2009 till mid-2010; so Aus­tralia re-lev­ered while the USA in par­tic­u­lar de-lev­ered. This ris­ing debt boosted aggre­gate demand after its ini­tial fall dur­ing the GFC, but the growth of debt is now at lev­els well below the pre-GFC peak.

Antic­i­pat­ing what might hap­pen from now on involves con­sid­er­ing one of the trick­i­est aspects of debt, its accel­er­a­tion. Since aggre­gate demand is income plus the change in debt, the change in aggre­gate demand is the change in income plus the accel­er­a­tion of debt. I define the “Credit Accel­er­a­tor” (ini­tially called the “Credit Impulse” by Biggs and Mayer 2010; Biggs, Mayer et al. 2010) as the ratio of the accel­er­a­tion of debt to GDP, and use this as an indi­ca­tion of how strongly the dynam­ics of pri­vate debt are going to impact upon eco­nomic per­for­mance and asset mar­kets.

Fig­ure 6 illus­trates why the “GFC” (which Amer­i­cans call the “Great Reces­sion”) was so much more dev­as­tat­ing in the USA than Aus­tralia: the decel­er­a­tion of debt was far more extreme, and lasted for much longer.

Fig­ure 6

Fig­ure 7 shows the rela­tion­ship between credit accel­er­a­tion and change in employ­ment in Aus­tralia since 2000. Though the dif­fer­ence between the mild Aus­tralian and the severe Amer­i­can down­turn was due to the much more severe decel­er­a­tion of debt in Amer­ica, pri­vate debt is now decel­er­at­ing in Aus­tralia, and the level of employ­ment is falling as a result.

Fig­ure 7 (Mea culpa–R^2 should be “Cor­re­la­tion Coef­fi­cient”)

A sim­i­lar phe­nom­e­non applies to the most impor­tant asset class in Aus­tralia, hous­ing. Mort­gage debt was the only com­po­nent of pri­vate debt to rise dur­ing the GFC—under the influ­ence of what I call the First Home Ven­dors Boost. Mort­gage debt is now decel­er­at­ing, and house prices are falling as a result. As expe­ri­ence has shown in the USA and Japan, this tends to be a run­away process, as falling house prices encour­age fur­ther falls in debt.

Fig­ure 8 (R^2 should be “Cor­re­la­tion Coef­fi­cient”)

Clearly, eco­nomic pol­icy is now far more com­plex than it appeared to be before the GFC. As we enter this Age of Delever­ag­ing, the worst thing we can do is apply poli­cies that appeared to work dur­ing the pre­ced­ing Age of Leverage—but were in fact pred­i­cated on ever-ris­ing pri­vate sec­tor indebt­ed­ness. Politi­cians should be scep­ti­cal of con­ven­tional eco­nomic advice at this time; it would be much wiser to study the his­tory of the 1930s instead.

Bernanke, B. S. (2000). Essays on the Great Depres­sion. Prince­ton, Prince­ton Uni­ver­sity Press.

Beze­mer, D. J. (2009). “No One Saw This Com­ing”: Under­stand­ing Finan­cial Cri­sis Through Account­ing Mod­els. Gronin­gen, The Nether­lands, Fac­ulty of Eco­nom­ics Uni­ver­sity of Gronin­gen.

Biggs, M. and T. Mayer (2010). “The Out­put Gap Conun­drum.” Intereconomics/Review of Euro­pean Eco­nomic Pol­icy
45(1): 11–16.

Biggs, M., T. Mayer, et al. (2010). “Credit and Eco­nomic Recov­ery: Demys­ti­fy­ing Phoenix Mir­a­cles.” SSRN eLi­brary.

Car­pen­ter, S. B. and S. Demi­ralp (2010). Money, Reserves, and the Trans­mis­sion of Mon­e­tary Pol­icy: Does the Money Mul­ti­plier Exist? Finance and Eco­nom­ics Dis­cus­sion Series. Wash­ing­ton, Fed­eral Reserve Board.

Fama, E. F. and K. R. French (2002). “Test­ing Trade-Off and Peck­ing Order Pre­dic­tions about Div­i­dends and Debt.” Review of Finan­cial Stud­ies
15(1): 1–33.

Fisher, I. (1933). “The Debt-Defla­tion The­ory of Great Depres­sions.” Econo­met­rica
1(4): 337–357.

Holmes, A. R. (1969). Oper­a­tional Con­straints on the Sta­bi­liza­tion of Money Sup­ply Growth. Con­trol­ling Mon­e­tary Aggre­gates. F. E. Mor­ris. Nan­tucket Island, The Fed­eral Reserve Bank of Boston: 65–77.

Keen, S. (2001). Debunk­ing eco­nom­ics: The naked emperor of the social sci­ences. Annan­dale Syd­ney & Lon­don UK, Pluto Press Aus­tralia & Zed Books UK.

Keen, S. (2011). Debunk­ing eco­nom­ics: The naked emperor dethroned? Lon­don, Zed Books.

Keen, S. (2011). “Debunk­ing Macro­eco­nom­ics.” Eco­nomic Analy­sis & Pol­icy
41(3): 147–167.

Koo, R. (2009). The Holy Grail of Macro­eco­nom­ics: Lessons from Japan’s Great Reces­sion. Wiley.

Koo, R. (2011). “The world in bal­ance sheet reces­sion: causes, cure, and pol­i­tics ” Real-World Eco­nom­ics Review
58(58): 19–37.

Krug­man, P. (2011). “Debt Is (Mostly) Money We Owe to Our­selves.” The Con­science of a Lib­eral 2012.

Krug­man, P. and G. B. Eggerts­son (2010). Debt, Delever­ag­ing, and the Liq­uid­ity Trap: A Fisher-Min­sky-Koo approach [2nd draft 2/14/2011]. New York, Fed­eral Reserve Bank of New York & Prince­ton Uni­ver­sity.

Kyd­land, F. E. and E. C. Prescott (1990). “Busi­ness Cycles: Real Facts and a Mon­e­tary Myth.” Fed­eral Reserve Bank of Min­neapo­lis Quar­terly Review
14(2): 3–18.

Min­sky, H. P. (1982). Can “it” hap­pen again? : essays on insta­bil­ity and finance. Armonk, N.Y., M.E. Sharpe.

Moore, B. J. (1979). “The Endoge­nous Money Stock.” Jour­nal of Post Key­ne­sian Eco­nom­ics
2(1): 49–70.

Schum­peter, J. A. (1934). The the­ory of eco­nomic devel­op­ment : an inquiry into prof­its, cap­i­tal, credit, inter­est and the busi­ness cycle. Cam­bridge, Mass­a­chu­setts, Har­vard Uni­ver­sity Press.


About Steve Keen

I am Professor of Economics and Head of Economics, History and Politics at Kingston University London, and a long time critic of conventional economic thought. As well as attacking mainstream thought in Debunking Economics, I am also developing an alternative dynamic approach to economic modelling. The key issue I am tackling here is the prospect for a debt-deflation on the back of the enormous private debts accumulated globally, and our very low rate of inflation.
Bookmark the permalink.