Euro­pean Dis­union and Endoge­nous Money

Flattr this!


Click here for this data in Excel: Debt­watchCfESI
Click here for this post in PDFDebt­watchCfESI

Steve Keen, Uni­ver­sity of West­ern Syd­ney
Matheus Gras­selli, Fields Insti­tute, Toronto

It ain’t what you don’t know that gets you into trou­ble. It’s what you know for sure that just ain’t so.” (Mark Twain)


That the Euro has fallen into cri­sis a mere decade after its intro­duc­tion is hardly sur­pris­ing. The intrin­sic prob­lems in its design were evi­dent to econ­o­mists as widely sep­a­rated intel­lec­tu­ally as Wynne God­ley and Mil­ton Fried­man. Writ­ing in 1992, God­ley observed that

The cen­tral idea of the Maas­tricht Treaty is that the EC coun­tries should move towards an eco­nomic and mon­e­tary union, with a sin­gle cur­rency man­aged by an inde­pen­dent cen­tral bank. But how is the rest of eco­nomic pol­icy to be run? As the treaty pro­poses no new insti­tu­tions other than a Euro­pean bank, its spon­sors must sup­pose that noth­ing more is needed. But this could only be cor­rect if mod­ern economies were self-adjust­ing sys­tems that didn’t need any man­age­ment at all.

I am dri­ven to the con­clu­sion that such a view – that economies are self-right­ing organ­isms which never under any cir­cum­stances need man­age­ment at all – did indeed deter­mine the way in which the Maas­tricht Treaty was framed… All that can legit­i­mately be done, accord­ing to this view, is to con­trol the money sup­ply and bal­ance the bud­get…

If a coun­try or region has no power to devalue, and if it is not the ben­e­fi­ciary of a sys­tem of fis­cal equal­i­sa­tion, then there is noth­ing to stop it suf­fer­ing a process of cumu­la­tive and ter­mi­nal decline lead­ing, in the end, to emi­gra­tion as the only alter­na­tive to poverty or star­va­tion. (God­ley 1992, pp. 3–4)

As the father of Mon­e­tarism (Fried­man 1969), Fried­man can be counted as one who did believe that “All that can legit­i­mately be done is to con­trol the money sup­ply and bal­ance the bud­get”. Yet in 1997, he too dis­par­aged the prospects for the Euro, because Europe did not meet the require­ments for a suc­cess­ful cur­rency union:

A com­mon cur­rency is an excel­lent mon­e­tary arrange­ment under some cir­cum­stances, a poor mon­e­tary arrange­ment under oth­ers…

Europe’s com­mon mar­ket exem­pli­fies a sit­u­a­tion that is unfa­vor­able to a com­mon cur­rency. It is com­posed of sep­a­rate nations, whose res­i­dents … have far greater loy­alty and attach­ment to their own coun­try than to … the idea of “Europe.”… goods move less freely than in the United States, and so does cap­i­tal. The Euro­pean Com­mis­sion … spends a small frac­tion of the total spent by gov­ern­ments in the mem­ber coun­tries. They, not the Euro­pean Union’s bureau­cra­cies, are the impor­tant polit­i­cal enti­ties… wages and prices in Europe are more rigid, and labor less mobile. In those cir­cum­stances, flex­i­ble exchange rates pro­vide an extremely use­ful adjust­ment mech­a­nism. (Fried­man 1997)

One might hope that, with econ­o­mists as dis­parate as God­ley & Fried­man able to agree that the Euro was always a bad idea, its unwinding—or at least the addi­tion of the miss­ing ele­ments from a gen­uine com­mon cur­rency, such as fis­cal union—could also be eas­ily achieved. But cur­rent events indi­cate that Europe may per­sist with its unsus­tain­able sys­tem until the social col­lapse and polit­i­cal dis­unity that God­ley and Fried­man antic­i­pated finally destroys per­haps even the EEC itself.

There are many polit­i­cal rea­sons for this dogged deter­mi­na­tion to fol­low folly to its end—including the dis­parate ben­e­fits and costs of the Euro to dif­fer­ent coun­tries, the inher­ent iner­tia in multi-coun­try polit­i­cal agree­ments, and Weimar Repub­lic-based fears that fis­cal expan­sion will cause hyper­in­fla­tion.

But eco­nomic the­ory has also played a role, by pro­mul­gat­ing the belief that gov­ern­ment debt should be con­strained, while pri­vate debt can be left unmon­i­tored. This belief was entrenched in the Maas­tricht Treaty’s lim­its on pub­lic debt, and is being enforced today via the selec­tive impo­si­tion of aus­ter­ity pro­grams on the weaker economies that are in breach of these lim­its.

Table 1: Selec­tive enforce­ment of Maas­tricht rules; coun­tries not in breach show in ital­ics (OCED Data)

Coun­try Debt 2010 Deficit 2012







































Slo­vak Repub­lic












The root of this belief is the Neo­clas­si­cal “exoge­nous” model of money cre­ation, in which “noth­ing is so unim­por­tant as the quan­tity of money expressed in terms of the nom­i­nal mon­e­tary unit” (Fried­man 1969, p. 1), the money sup­ply is deter­mined by the actions of the Cen­tral Bank, the level of pri­vate debt is deter­mined by the sup­ply of and demand for loan­able funds, and pri­vate banks, as mere inter­me­di­aries between savers and bor­row­ers in the loan­able funds mar­ket, can be ignored in macro­eco­nom­ics.

Iron­i­cally, though belief in this vision of money played a major role in the design of the Euro­pean Mon­e­tary Union, and it is accepted by Neo­clas­si­cal econ­o­mists who argue for aus­ter­ity as the solu­tion to the cri­sis, it is also adhered to by Neo­clas­si­cal crit­ics of aus­ter­ity. With unabashed hubris, this lat­ter group—the self-described New Keynesians—are now claim­ing to be vin­di­cated by the cri­sis, while appro­pri­at­ing Hyman Min­sky and cast­ing them­selves as the future of eco­nom­ics in the process:

We econ­o­mists who are steeped in eco­nomic and finan­cial his­tory – and aware of the his­tory of eco­nomic thought con­cern­ing finan­cial crises and their effects – have rea­son to be proud of our analy­ses over the past five years…

So the big les­son is sim­ple: trust those who work in the tra­di­tion of Wal­ter Bage­hot, Hyman Min­sky, and Charles Kindle­berger. That means trust­ing econ­o­mists like Paul Krug­man, Paul Romer, Gary Gor­ton, Car­men Rein­hart, Ken Rogoff, Raghu­ram Rajan, Larry Sum­mers, Barry Eichen­green, Olivier Blan­chard, and their peers. Just as they got the recent past right, so they are the ones most likely to get the dis­tri­b­u­tion of pos­si­ble futures right. (deLong 2012)

The revi­sion­ism in this argu­ment is breath­tak­ing. As late as Sep­tem­ber 2009, Blan­chard was pro­claim­ing that “the stage of macro [by which he meant DSGE mod­el­ing] is good” (O. J. Blan­chard 2008, p. 210; O. Blan­chard 2009, p. 2). Krug­man, described here as work­ing “in the tra­di­tion of Min­sky” first actu­ally read Min­sky in May 2009 (Krug­man 2009)—and found what he read (Min­sky 1986) rather dif­fi­cult to fathom. And non-Neo­clas­si­cal experts on Min­sky with a long pedigree—like Randy Wray (Larry Ran­dall Wray 1988) and myself (Keen 1995)—were com­pletely ignored.

Though it is laugh­able, this revi­sion­ism should not be treated lightly. IS-LM analy­sis, which those who are truly aware of the his­tory of eco­nomic thought know was a Wal­rasian model mas­querad­ing as Keynes (J. R. Hicks 1935; J. Hicks 1981), was orig­i­nally devel­oped in what pur­ported to be a book review (J. R. Hicks 1937) of The Gen­eral The­ory (Keynes 1936). There is thus a prospect that a dis­as­ter caused in large mea­sure by impos­ing a Neo­clas­si­cal fan­tasy on the real world (the belief in self-equi­li­brat­ing mar­kets) and which is being ampli­fied by impos­ing yet another Neo­clas­si­cal fan­tasy (aus­ter­ity as an eco­nomic stim­u­lus) will be used to entrench still a third Neo­clas­si­cal fan­tasy as the pin­na­cle of post-cri­sis macro­eco­nom­ics (exoge­nous money and a neo-Wal­rasian macro­eco­nom­ics derived from it by adding in a pinch of imper­fect com­pe­ti­tion and incom­plete infor­ma­tion).

While there is no cer­tainty that this Neo­clas­si­cal jug­ger­naut can be stopped, it is cer­tain that it will roll over non-Neo­clas­si­cal oppo­si­tion unless a coher­ent, empir­i­cally accu­rate, and com­pelling alter­na­tive the­ory is devel­oped. In what fol­lows I set out essen­tial ele­ments of that alternative—starting with the empir­i­cal point of diver­gence between the dom­i­nant Neo­clas­si­cal expla­na­tion for the sever­ity and longevity of this cri­sis, and an endoge­nous money expla­na­tion.

Private Debt and Aggregate Demand

The New Keynesian Analysis: a “Liquidity Trap”

The New Key­ne­sian expla­na­tion for the cri­sis is that we are in a liq­uid­ity trap as defined by Hicks (J. R. Hicks 1937, pp. 154–155), when the capac­ity of mon­e­tary pol­icy to stim­u­late the econ­omy fails:

For the last fifty years the busi­ness of end­ing reces­sions has basi­cally been the job of the Fed­eral Reserve, which … con­trols the “mon­e­tary base,” … Well, the Fed has tripled the size of the mon­e­tary base since 2008; yet the econ­omy remains depressed… What the Fed can do by push­ing more cash into the banks is drive down inter­est rates… But … it can’t push them below zero… Unfor­tu­nately, a zero rate turned out not to be low enough, because the burst­ing of the hous­ing bub­ble had done so much dam­age… And that’s the liq­uid­ity trap: it’s what hap­pens when zero isn’t low enough. (Krug­man 2012f, pp. 31–33)

Krug­man asserts that this alters an essen­tial reg­u­lar­ity in macro­eco­nom­ics. Nor­mally, the level and rate of change of pri­vate debt are irrel­e­vant to macro­eco­nom­ics, because pri­vate debt is sim­ply a trans­fer from one agent to another, with a net sum impact on the macro­econ­omy of very nearly zero:

to a first approx­i­ma­tion debt is money we owe to our­selves… the over­all level of debt makes no dif­fer­ence to aggre­gate net worth—one person’s lia­bil­ity is another person’s asset. It fol­lows that the level of debt mat­ters only if the dis­tri­b­u­tion of net worth mat­ters, if highly indebted play­ers face dif­fer­ent con­straints from play­ers with low debt.… when debt is ris­ing, it’s not the econ­omy as a whole bor­row­ing more money. It is, rather, a case of less patient peo­ple … bor­row­ing from more patient peo­ple. The main limit on this kind of bor­row­ing is the con­cern of those patient lenders about whether they will be repaid, which sets some kind of ceil­ing on each individual’s abil­ity to bor­row. (Krug­man 2012f, pp. 146–147, empha­sis added)

The cri­sis occurred—and pri­vate debt assumed macro­eco­nomic significance—because “for some rea­son” (Krug­man and Eggerts­son 2010, p. 3), this ceil­ing was reduced. The asym­met­ric con­se­quences this had for spend­ing is why debt mat­ters: because decreased spend­ing by indebted indi­vid­u­als was not off­set by ris­ing spend­ing by savers. This itself could have been coun­tered had the Fed­eral Reserve had the flex­i­bil­ity needed to drop inter­est rates, but nom­i­nal rates were already his­tor­i­cally low, and the Fed­eral Reserve rapidly came up against the zero lower bound:

What hap­pened in 2008 was a sud­den down­ward revi­sion of those ceil­ings. This down­ward revi­sion has forced the debtors to pay down their debt, rapidly, which means spend­ing much less. And the prob­lem is that the cred­i­tors don’t face any equiv­a­lent incen­tive to spend more… because of the sever­ity of the “delever­ag­ing shock,” even a zero inter­est rate isn’t low enough to get them to fill the hole left by the col­lapse in debtors’ demand… (Krug­man 2012f, p. 147)

This in turn means that fis­cal pol­icy must be used to sup­port mon­e­tary pol­icy, so that ris­ing pub­lic-sec­tor spend­ing can fill the gap between exces­sive delever­ag­ing by bor­row­ers and insuf­fi­cient addi­tional spend­ing by savers:

we had a period of too much opti­mism about debt, in which debtors bor­rowed and spent too much; since one person’s debt is another’s asset, cred­i­tors had to be induced to spend less via high real inter­est rates. Then peo­ple remem­bered the dan­gers of debt, and we moved from lever­ag­ing to delever­ag­ing. But the prob­lem is that this isn’t sym­met­ric, because you can’t get real inter­est rates low enough to induce suf­fi­cient spend­ing on the part of those not deep in debt.

So one way to explain our depres­sion is to say that debtors, as a group, are try­ing to delever­age too fast, in the sense that the col­lec­tive rate at which they are try­ing to pay down debt isn’t fea­si­ble given the zero lower bound on inter­est rates.

And that’s the role of fis­cal pol­icy: its goal is not to stop aggre­gate delever­ag­ing … but to slow it down to a pace that can be accom­mo­dated by mon­e­tary pol­icy.(Krug­man 2012e)

Like almost all Post Key­ne­sian econ­o­mists, I sup­port gov­ern­ment deficit spend­ing dur­ing this cri­sis (though I argue it alone is not enough). But Krugman’s argu­ment about why deficits are needed is strongly fal­si­fied by the empir­i­cal data.

Testing the “Liquidity Trap”

Two key empir­i­cal propo­si­tions can be derived from Krugman’s analy­sis:

  1. High real inter­est rates should be cor­re­lated with high rates of growth of real debt; and
  2. The level and rate of change of debt should only mat­ter when the zero lower bound causes a liq­uid­ity trap

The first propo­si­tion is mod­er­ately sup­ported by the data. Fig­ure 1 shows the two fac­tors deter­min­ing the real Reserve inter­est rate: the Fed­eral Reserve Funds rate and the infla­tion rate.

Fig­ure 1: Fed­eral Funds Rate and Infla­tion

Fig­ure 2 shows the cor­re­la­tion between the real Reserve rate and the infla­tion-adjusted change in the level of pri­vate debt (derived from the Flow of Funds Table L.1). It is a rea­son­ably strong 0.46 across the whole time series, but only 0.21 for the period prior to the Zero Lower Bound (the ZLB—defined as when the Fed­eral Fund tar­get rate dropped to 0.125 per cent in Jan­u­ary 2009). Most of the strength of the cor­re­la­tion comes from after the ZLB was reached, when the cor­re­la­tion rises to 0.65. Krugman’s first propo­si­tion is thus at best only weakly sup­ported by the data.

Fig­ure 2: Real inter­est rates and pri­vate debt growth

His sec­ond proposition—that pri­vate debt should have no macro­eco­nomic sig­nif­i­cance except after the ZLB—is strongly con­tra­dicted by the data. Firstly, the neg­a­tive cor­re­la­tion that Krug­man expected to apply between change in debt and unem­ploy­ment after the ZLB is there, but it is weak—only –0.27. How­ever, his expec­ta­tion that there would be no rela­tion­ship between the change in debt and unem­ploy­ment prior to the ZLB is wrong: the cor­re­la­tion between changes in real pri­vate debt and unem­ploy­ment is sub­stan­tially stronger before the ZLB: –0.69. Over the whole time period, the cor­re­la­tion is –0.55.

Krugman’s “liq­uid­ity trap” expla­na­tion for why debt mat­ters can thus com­fort­ably be rejected: changes in the aggre­gate level of debt have macro­eco­nomic sig­nif­i­cance at all times, even when there is clearly no liq­uid­ity trap.

Fig­ure 3: Real change in pri­vate debt and unem­ploy­ment

The fail­ure of the “liq­uid­ity trap” hypoth­e­sis as to and when why pri­vate debt mat­ters, paves the way to test an alter­na­tive: that pri­vate debt always mat­ters, because changes in pri­vate debt add to aggre­gate demand.

Testing Endogenous Money

The propo­si­tion that changes in pri­vate debt add to aggre­gate demand sparked Krugman’s blo­gos­phere stoush with me (Keen, 2012; Krug­man 2012d, 2012c, 2012g, 2012a, 2012b) in early 2012:

Keen then goes on to assert that lend­ing is, by def­i­n­i­tion (at least as I under­stand it), an addi­tion to aggre­gate demand. I guess I don’t get that at all. If I decide to cut back on my spend­ing and stash the funds in a bank, which lends them out to some­one else, this doesn’t have to rep­re­sent a net increase in demand. Yes, in some (many) cases lend­ing is asso­ci­ated with higher demand, because resources are being trans­ferred to peo­ple with a higher propen­sity to spend; but Keen seems to be say­ing some­thing else, and I’m not sure what. I think it has some­thing to do with the notion that cre­at­ing money = cre­at­ing demand, but again that isn’t right in any model I under­stand. (Krug­man 2012c)

To test this propo­si­tion, the change in debt needs to be com­pared to GDP, since we are mea­sur­ing its con­tri­bu­tion to aggre­gate demand, and a 10% change in debt rep­re­sented a 10% increase in demand above GDP in 1970 (when the debt to GDP ratio was 100%) but a 30% increase in 2009 (when the ratio peaked at 300%)—see Fig­ure 4.

Fig­ure 4: The impact of a given per­cent­age change in debt has grown as debt has grown rel­a­tive to GDP

Fig­ure 5 shows that the empir­i­cal rela­tion­ship between unem­ploy­ment and the change in pri­vate debt (mea­sured as a per­cent­age of GDP) is high and rel­a­tively sta­ble either side of the ZLB. It is –0.76 for the entire time period from Jan­u­ary 1980 till March 2012, –0.63 since the ZLB, and –0.84 prior to the ZLB.

Fig­ure 5: Change in pri­vate debt and unem­ploy­ment

Table 2 sum­ma­rizes the cor­re­la­tions uncov­ered here. Clearly, there is a strong empir­i­cal reg­u­lar­ity between change in debt and macro­eco­nomic per­for­mance that sur­vives peri­ods of both boom and bust, and low and high real and nom­i­nal inter­est rates.

Table 2: Sum­mary of cor­re­la­tions

Cor­re­la­tion Coef­fi­cients Before ZLB After ZLB 1980–2012
Real inter­est rate & change in real pri­vate debt p.a. 0.21 0.65 0.46
Change in real pri­vate debt & unem­ploy­ment –0.69 –0.27 –0.55
Change in debt per­cent of GDP & unem­ploy­ment –0.84 –0.63 –0.76

These results call for a hypoth­e­sis that is con­sis­tent with them: the endoge­nous money hypoth­e­sis.

Endogenous Money

The essence of endoge­nous money hypoth­e­sis is that banks cre­ate spend­ing power for bor­row­ers with­out reduc­ing the spend­ing power of savers. If true, this makes banks far more than mere inter­me­di­aries, and a cru­cial part of a valid the­ory of macro­eco­nom­ics. An essen­tial pre-req­ui­site of this the­ory is that bank lend­ing is not effec­tively con­strained by the Cen­tral Bank.

Both the essence and the pre-req­ui­site receive sub­stan­tial sup­port from exist­ing empir­i­cal research (as well as the results above), and from infor­ma­tion on the actual prac­tices of Cen­tral Banks.

Empirical Data

While test­ing the “peck­ing order” the­ory of cor­po­rate div­i­dends, Fama and French uncov­ered a strong rela­tion­ship between changes in aggre­gate cor­po­rate debt and the level of cor­po­rate invest­ment. Using the Com­pu­s­tat data­base of com­pany reports from pub­licly-traded US non-finan­cial cor­po­ra­tions between 1951 & 1996, Fama and French cal­cu­lated aggre­gate non-finan­cial cor­po­rate invest­ment, and cor­re­lated it with equity issue, retained earn­ings, and new debt (see Fig­ure 6).

Fig­ure 6: Cor­re­la­tions of invest­ment to new equity, retained earn­ings and new debt (Fama & French 1999, p. 1954)

They con­cluded that “the source of financ­ing most cor­re­lated with invest­ment is long-term debt”:

new net issues of stock do not move closely with invest­ment… the cor­re­la­tion of invest­ment, It, and new net issues of stock, dSt, is only 0.19… retained cash earn­ings move more closely with invest­ment. The cor­re­la­tion between It and RCEt is indeed higher, 0.56, but far from per­fect. The source of financ­ing most cor­re­lated with invest­ment is long-term debt. The cor­re­la­tion between It and dLTDt is 0.79…These cor­re­la­tions con­firm the impres­sion from Fig­ure 3 that debt plays a key role in accom­mo­dat­ing year-by-year vari­a­tion in invest­ment. (Fama and French 1999, p. 1954)


Kyd­land and Prescott devel­oped a novel method for test­ing eco­nomic mod­els, and were can­did that their results con­tra­dicted Neo­clas­si­cal the­ory:

The pur­pose of this arti­cle is to present the busi­ness cycle facts in light of estab­lished neo­clas­si­cal growth the­ory… We find these fea­tures inter­est­ing because the pat­terns they seem to dis­play are incon­sis­tent with the the­ory. (Kyd­land and Prescott 1990, p. 4)

On the rela­tion­ship between reserves and credit, they con­cluded that:

the mon­e­tary base lags the cycle slightly… The dif­fer­ence of M2-M1 leads the cycle by … about three quar­ters… The fact that the trans­ac­tion com­po­nent of real cash bal­ances (M1) moves con­tem­po­ra­ne­ously with the cycle while the much larger non­trans­ac­tion com­po­nent (M2) leads the cycle sug­gests that credit arrange­ments could play a sig­nif­i­cant role in future busi­ness cycle the­ory… (Kyd­land and Prescott 1990, p. 15)

These find­ings con­firmed the results of the Post Key­ne­sian econ­o­mist Basil Moore (Moore 1979, 1983, 1988), that bank lend­ing was not reserve-con­strained. The “Hor­i­zon­tal­ist” model he devised (Moore 1988, 1997) sparked the devel­op­ment of the endoge­nous money hypoth­e­sis by Post Key­ne­sian econ­o­mists.

Practical issues

The tim­ing rela­tion between bank loans and deposits and reserves is the oppo­site of what would be required for reserves to act as a con­straint on lend­ing. Fed­eral Reserve Bank of New York Vice Pres­i­dent Alan Holmes put it this way when express­ing his oppo­si­tion to Mon­e­tarism in 1969:

The idea of a reg­u­lar injec­tion of reserves … also suf­fers from a naive assump­tion that the bank­ing sys­tem only expands loans after the Sys­tem (or mar­ket fac­tors) have put reserves in the bank­ing sys­tem. In the real world, banks extend credit, cre­at­ing deposits in the process, and look for the reserves later… the reserves required to be main­tained by the bank­ing sys­tem are pre­de­ter­mined by the level of deposits exist­ing two weeks ear­lier. (Holmes 1969, p. 73)

The rela­tion­ship that loans and deposits lead and reserves lag is more pro­nounced today, with the reserve lag in the USA now being 30 days (O’Brien 2007, Table 12, p. 52). The Euro­pean Cen­tral Bank has also recently con­firmed that the Post Key­ne­sian posi­tion that “loans cre­ate deposits, and deter­mine reserves with a lag” accu­rately describes pri­vate and Cen­tral Bank pro­ce­dures in the EU:

In fact, the ECB’s reserve require­ments are back­ward-look­ing, i.e. they depend on the stock of deposits (and other lia­bil­i­ties of credit insti­tu­tions) sub­ject to reserve require­ments as it stood in the pre­vi­ous period, and thus after banks have extended the credit demanded by their cus­tomers. (ECB 2012, p. 21, empha­sis added)

Lastly, the col­lapse in the ratio of broad money to base money dur­ing and after the cri­sis inspired an FRB Dis­cus­sion Paper which con­cluded that the “money mul­ti­plier” model of money cre­ation, which is an intrin­sic part of the “exoge­nous money hypoth­e­sis”, is false:

the rela­tion­ships implied by the money mul­ti­plier do not exist in the data for the most liq­uid and well-cap­i­tal­ized banks. Changes in reserves are unre­lated to changes in lend­ing, and open mar­ket oper­a­tions do not have a direct impact on lend­ing. We con­clude that the text­book treat­ment of money in the trans­mis­sion mech­a­nism can be rejected. (Car­pen­ter and Demi­ralp 2010, pp. 27–28)

Theoretical arguments

Though the con­cept that changes in the level of pri­vate debt affect aggre­gate demand had sev­eral prog­en­i­tors, the direct antecedent of today’s endoge­nous money the­o­rists was Joseph Schum­peter. In try­ing to explain the ori­gin of prof­its (while start­ing from a Wal­rasian gen­eral equi­lib­rium in which prof­its were zero), Schum­peter linked dis­con­tin­u­ous change by entre­pre­neurs with endoge­nous money cre­ation by banks. He argued that an entre­pre­neur was the pos­ses­sor of an idea that could trans­form pro­duc­tion who lacked the mon­e­tary means to put that idea into oper­a­tion (Schum­peter acknowl­edged but down­played the pos­si­bil­ity of invest­ing out of retained earn­ings). Lack­ing rev­enue from sell­ing goods in “the cir­cu­lar flow”, to acquire money the entre­pre­neur had to get a bank loan:

the entrepreneur—in prin­ci­ple and as a rule—does need credit … in order to pro­duce at all… And this pur­chas­ing power does not flow towards him auto­mat­i­cally, as to the pro­ducer in the cir­cu­lar flow, by the sale of what he pro­duced in pre­ced­ing peri­ods… he must bor­row it… He can only become an entre­pre­neur by pre­vi­ously becom­ing a debtor… He is the typ­i­cal debtor in cap­i­tal­ist soci­ety. (Schum­peter 1934, p. 102).

This bank loan in turn did not require a saver to forego spend­ing power, because banks could cre­ate addi­tional spend­ing power sim­ply by dou­ble-entry book­keep­ing:

Even though the con­ven­tional answer to our ques­tion is not obvi­ously absurd, yet there is another method of obtain­ing money for this pur­pose … the cre­ation of pur­chas­ing power by banks… 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, pp. 72–73).

This meant that total demand (which Schum­peter con­fus­ingly describes as “total credit”) in a grow­ing cap­i­tal­ist econ­omy exceeds demand from the sale of goods and ser­vices alone (which he con­fus­ingly describes as “fully cov­ered credit”, in con­trast to “abnor­mal credit”—which isn’t abnor­mal at all!—that emanates from bank loans to entre­pre­neurs):

From this it fol­lows, there­fore, that in real life total credit must be greater than it could be if there were only fully cov­ered credit. The credit struc­ture projects … beyond the exist­ing com­mod­ity basis… Nor­mal credit cre­ates claims to the social div­i­dend, which rep­re­sent … pre­vi­ous deliv­ery of exist­ing goods. That kind of credit, which is des­ig­nated by tra­di­tional opin­ion as abnor­mal, also cre­ates claims to the social prod­uct, which … could only be described as cer­tifi­cates of future ser­vices or of goods yet to be pro­duced. Thus there is a fun­da­men­tal dif­fer­ence between the two cat­e­gories, in their nature as well as in their effects. Both serve the same pur­pose as means of pay­ment and are exter­nally indis­tin­guish­able. But the one embraces means of pay­ment to which there is a cor­re­spond­ing con­tri­bu­tion to the social prod­uct, the other means of pay­ment to which so far noth­ing cor­re­sponds…. (Schum­peter 1934, pp. 101–102).

In rather clearer lan­guage, Min­sky elab­o­rated Schumpeter’s key point—that aggre­gate demand in a grow­ing cap­i­tal­ist econ­omy exceeds income, and the dif­fer­ence is made up by ris­ing debt:

If income is to grow, the finan­cial mar­kets, where the var­i­ous plans to save and invest are rec­on­ciled, must gen­er­ate an aggre­gate demand that, aside from brief inter­vals, is ever ris­ing. For real aggre­gate demand to be increas­ing, … it is nec­es­sary that cur­rent spend­ing plans, summed over all sec­tors, be greater than cur­rent received income and that some mar­ket tech­nique exist by which aggre­gate spend­ing in excess of aggre­gate antic­i­pated income can be financed. It fol­lows that over a period dur­ing which eco­nomic growth takes place, at least some sec­tors finance a part of their spend­ing by emit­ting debt or sell­ing assets. (Min­sky 1982, p. 6, empha­sis added)

Min­sky added that money was bor­rowed not only by Schum­peter­ian entre­pre­neurs, whose spend­ing would ulti­mately increase society’s pro­duc­tive capac­ity, but also by Ponzi Financiers, who would drive up asset prices and finan­cial fragility:

A Ponzi finance unit is a spec­u­la­tive financ­ing unit for which the income com­po­nent of the near term cash flows falls short of the near term inter­est pay­ments on debt so that for some time in the future the out­stand­ing debt will grow due to inter­est on exist­ing debt… Ponzi units can ful­fill their pay­ment com­mit­ments on debts only by bor­row­ing (or dis­pos­ing of assets)… a Ponzi unit must increase its out­stand­ing debts. (Min­sky 1982, p. 24)

There are there­fore two sources of mon­e­tary aggre­gate demand—income gained from sell­ing goods and ser­vices, and new credit cre­ated by debt issued by banks—and two broad classes of expenditure—newly pro­duced goods and ser­vices, and finan­cial claims on exist­ing assets. This can be sum­ma­rized in the fol­low­ing state­ment:

Aggre­gate Demand equals income plus the change in debt, and Aggre­gate Sup­ply equals Out­put plus new pur­chases of finan­cial assets.

The ini­tial reac­tion of most econ­o­mists to this propo­si­tion is that it must involve dou­ble-count­ing. In the next sec­tion, I resort to math­e­mat­i­cal logic to show that this is not so.

Mathematical Proof

Econ­o­mists cus­tom­ar­ily think in “dis­crete time” when writ­ing equa­tions; this sec­tion will be eas­ier to fol­low if the reader instead thinks in “con­tin­u­ous time”—so that any term of the form X implies “the value of X at time t” while rates of change expres­sions refer to the change in a vari­able at an instant in time. Through­out I will use for income, for expen­di­ture, W for Wages, for profit, and D for debt. I begin with a sim­ple model in which all wages are con­sumed, and cap­i­tal­ists can both con­sume and invest but do not spec­u­late (i.e., the pur­chas­ing of exist­ing assets is not con­sid­ered), and pro­gres­sively con­sider more com­plex arrange­ments.

Closed Economy, no government or speculation

We start with the iden­tity that total income equals wages plus prof­its:

Prof­its are either con­sumed or retained for invest­ment:

Expen­di­ture is either on con­sumer goods or invest­ment goods:

Con­sump­tion is financed by wages and retained earn­ings, so that the sum expended buy­ing con­sumer goods equals wages plus retained earn­ings:

Invest­ment, as Fama and French’s empir­i­cal research con­firmed, is financed by invest­ment plus the change in debt:

Expen­di­ture and income are there­fore:

Sub­tract­ing income from expen­di­ture yields the result that the dif­fer­ence between expen­di­ture and income equals the change in debt—which is the argu­ment put by both Schum­peter and Min­sky:

There­fore aggre­gate demand equals income plus the change in debt:

Closed economy with speculation by capitalists

Dis­trib­uted prof­its are now either con­sumed or used for spec­u­la­tion:

Expen­di­ture is on both goods and ser­vices and assets:

Pur­chases of assets are financed out of prof­its and money bor­rowed for spec­u­la­tion:

Two types of debt are now con­sid­ered: debt for invest­ment and debt for spec­u­la­tion . Expen­di­ture and income are now respec­tively:

Sub­tract­ing from yields:

Closed economy with speculation and government

Expen­di­ture now includes net gov­ern­ment spend­ing :

Gov­ern­ment spend­ing equals the change in gov­ern­ment debt:

Expen­di­ture is thus equal to income plus the sum of the change in invest­ment, spec­u­la­tive and gov­ern­ment debt:

Though the maths is easy to fol­low, I am aware that the argu­ment still appears to involve dou­ble-count­ing to many econ­o­mists, so some fur­ther clar­i­fi­ca­tion is in order. I believe that this con­fu­sion arises from a ten­dency to think about this rela­tion­ship in an “ex post” manner—comparing recorded expen­di­ture with recorded income. This com­par­i­son is valid for record­ing what has hap­pened, but not valid for con­sid­er­ing the causal process in which, as Keynes empha­sized, expen­di­ture pre­cedes income, and the income for invest­ment is in part bor­rowed:

I pro­ceed to the third pos­si­ble source of con­fu­sion, due to the fact (which may deserve more empha­sis than I have given it pre­vi­ously) that an invest­ment deci­sion (Prof. Ohlin’s invest­ment ex-ante) may some­times involve a tem­po­rary demand for money before it is car­ried out, quite dis­tinct from the demand for active bal­ances which will arise as a result of the invest­ment activ­ity whilst it is going on. This demand may arise in the fol­low­ing way.

Planned investment—i.e. invest­ment ex-ante—may have to secure its ” finan­cial pro­vi­sion ” before the invest­ment takes place…There has, there­fore, to be a tech­nique to bridge this gap between the time when the deci­sion to invest is taken and the time when the cor­rel­a­tive invest­ment and sav­ing actu­ally occur. This ser­vice may be pro­vided either by the new issue mar­ket or by the banks;—which it is, makes no dif­fer­ence.” (Keynes 1937, p. 246)

A styl­ized flow dia­gram might also help. Incomes from “the cir­cu­lar flow” are gen­er­ated by the turnover of the exist­ing stock of money, while newly bor­rowed money adds spend­ing power of an equiv­a­lent amount when it is spent into the econ­omy, and also boosts the amount of money in circulation—thus enabling incomes to grow as well.

Fig­ure 7: A styl­ized cir­cu­lar flow plus new debt dia­gram

Finally, con­sider the rela­tion­ship at a moment in time . Wage income for that instant will be , while prof­its are and the change in debt is . Total spend­ing between two instants t1 and t2 is thus the inte­gral of these flows over that time period (where all 3 can change in mag­ni­tude over the inter­val, and influ­ence each other as well:

The change in expen­di­ture between two time peri­ods will there­fore be the dif­fer­ences in wages and prof­its plus the dif­fer­ences in the rate of change of debt:

This in turn means that the change in expen­di­ture equals the change in incomes plus the accel­er­a­tion of debt—a point first noted by Biggs, Meyer and Pick (Biggs et al. 2010; Biggs and Mayer 2010), when they chris­tened the accel­er­a­tion in debt (divided by GDP) as the “Credit Impulse”. I pre­fer to call the “Credit Accel­er­a­tor” to empha­size that it is not an occa­sional and tran­sient phe­nom­e­non, but a per­ma­nent fea­ture of a credit-based econ­omy.

This leaves one last issue: what is the mech­a­nism by which banks endoge­nously cre­ate credit with­out being con­strained by exist­ing sav­ings or the reserve require­ments set by the Cen­tral Bank?

The Mechanism

Neo­clas­si­cal econ­o­mists accept that the Cen­tral Bank has the capac­ity to cre­ate money “out of thin air”:

The cru­cial thing is where the Fed gets the funds with which it pur­chases assets. And the answer is that it cre­ates them out of thin air… there’s noth­ing behind that credit; the Fed has the unique right to con­jure money into exis­tence when­ever it chooses. (Krug­man 2012f, p. 153)

This unique right is the essen­tially infi­nite licence to cre­ate the cur­rency that a Cen­tral Bank of a sov­er­eign nation possesses—or rather has con­ferred upon it by the gov­ern­ment. But Neo­clas­si­cal econ­o­mists ignore the fact that the pos­ses­sion of a pri­vate bank­ing licence con­fers a similar—though not infinite—capacity to a bank to cre­ate money “out of thin air”. The exer­cise of both these capac­i­ties deter­mines the money sup­ply in our mixed state-credit finan­cial sys­tem.

The mech­a­nism, in both cases, com­mences with the capac­ity to cre­ate match­ing assets and lia­bil­i­ties via dou­ble-entry book­keep­ing, as “Mod­ern Mon­e­tary The­ory” (MMT) econ­o­mists empha­size:

Credit and debt are two sides of the same coin. Both cred­i­tor and debtor are sin­ful. They bal­ance. Exactly. The sin­ful bal­ance is ensured by dou­ble-entry book-keep­ing. (L. R. Wray 2011)

In its model of gov­ern­ment money cre­ation, MMT has to date treated the gov­ern­ment and the cen­tral bank as a sin­gle entity for the sake of sim­plic­ity (Full­wiler et al. 2012).

it is con­ve­nient to con­sol­i­date the trea­sury and cen­tral bank accounts into a “gov­ern­ment account”. To be sure, the real world is more com­pli­cated… But, as Paul David­son has fre­quently noted, the appro­pri­ate gen­eral case is the one that makes the fewest assump­tions while enabling analy­sis or under­stand­ing of the fun­da­men­tal or “true” nature of the object of inquiry. (Full­wiler et al. 2012, p. 20)

This con­sol­i­da­tion has been sub­ject to crit­i­cism by other Post Key­ne­sians (Fiebiger 2012). I will start with a dou­ble-entry book­keep­ing rep­re­sen­ta­tion of this arguably over-sim­pli­fied model—and an equally over-sim­pli­fied model of pri­vate banking—before mov­ing on to a more gen­eral rep­re­sen­ta­tion of both.

Simplified double-entry model of mixed state-credit money creation

Kel­ton, Full­wiler & Wray describe their sim­plest model in the fol­low­ing terms:

con­sider a con­sol­i­dated gov­ern­ment (cen­tral bank plus trea­sury) run­ning a deficit. The basic trans­ac­tions could be listed as the fol­low­ing:

1A. The government’s spend­ing cred­its bank accounts with reserve bal­ances (HPM). These accounts are lia­bil­i­ties of the government/central bank.

2A. Banks credit the deposit accounts of the spend­ing recip­i­ents. So, over­all, the increased reserve bal­ances have raised bank assets while the increased deposits have increased bank lia­bil­i­ties by the same amount. (Full­wiler et al. 2012, p. 21)

The sim­plest dou­ble-entry-con­sis­tent ren­di­tion of this model that I can develop is shown in Fig­ure 8 and Fig­ure 9. The capac­ity of a sov­er­eign gov­ern­ment to cre­ate its own cur­rency is shown as the intan­gi­ble asset of a “Fiat Licence”, with an ini­tial value that is essen­tially lim­it­less. That in turn lets it cre­ate the lia­bil­ity of its cur­rency, which is stored in a “Cen­tral Bank Vault”.

Gov­ern­ment spend­ing begins with a trans­fer from this Vault to the Treasury’s Deposit account (labeled row 1 in Fig­ure 8), and its record­ing as a trans­fer of Assets from the Fiat Licence to the Treasury’s Loan account (row 2). Spend­ing of this bor­rowed money into the econ­omy is shown on row 3 of Fig­ure 8 and row 1 of Fig­ure 9. The reverse oper­a­tions of tax­a­tion are shown on rows 4–6 of Fig­ure 8 and row 2 of Fig­ure 9.

The pri­vate sector’s money-cre­ation process is shown on rows 3 and 4 of Fig­ure 9.

Fig­ure 8: Gov­ern­ment com­po­nent of the sim­plest MMT money cre­ation model

Fig­ure 9: Pri­vate bank com­po­nent of the sim­plest MMT money cre­ation model

Since all these oper­a­tions are flows, the sym­bolic sum of each col­umn rep­re­sents the rate of change of the account shown in that col­umn. The cru­cial flow equa­tions are the fol­low­ing:

Two phe­nom­ena are evi­dent in this highly sim­pli­fied model:

  1. The total amount of money in cir­cu­la­tion in the econ­omy is increased by gov­ern­ment deficits, and by the pri­vate bank­ing sector’s loans to the non-bank pub­lic (when they exceed the non-bank public’s repay­ments of those loans).
  2. Gov­ern­ment deficit oper­a­tions affect pri­vate bank reserves, but pri­vate bank reserves play no role in pri­vate bank lend­ing.

How­ever, these results could be arti­facts of the exces­sively sim­pli­fied nature of the model. I now con­sider a more com­plete (but by no means full and final) model in which:

  1. The Cen­tral Bank and Trea­sury are sep­a­rated;
  2. Trea­sury must sell bonds to bond pur­chasers to raise rev­enue;
  3. The Cen­tral Bank can at its dis­cre­tion pur­chase gov­ern­ment bonds of pri­vate bond hold­ers; and
  4. Pri­vate banks lend from reserves.

More complete double-entry model of mixed state-credit money creation

Three accounts are now nec­es­sary:

  1. The Trea­sury;
  2. The Cen­tral Bank; and
  3. The Pri­vate Bank Sec­tor

The Trea­sury has the intan­gi­ble asset of a licence to cre­ate gov­ern­ment bonds, which is matched by the lia­bil­ity of the bonds it cre­ates. Hav­ing planned its spend­ing, it then sells bonds to bond dealers—and these bonds can in turn be pur­chased either by the non-bank pub­lic, or by the Cen­tral Bank.

Fig­ure 10: The Trea­sury

Hav­ing sold those bonds, the Trea­sury now has money in its deposit account at the Cen­tral Bank, from which it can spend. As before, this spend­ing increases the reserves of the pri­vate banks (which are a lia­bil­ity of the Cen­tral Bank) and increases the non-bank public’s deposits at the pri­vate banks by the same amount.

Fig­ure 11: The Cen­tral Bank

The flow of loans by the pri­vate bank­ing sec­tor to the non-bank pub­lic is mod­eled as a trans­fer of the bank­ing sector’s assets from reserves to loans, matched by the pri­vate non-bank sec­tor deposit­ing the flow in its deposit accounts. The two oper­a­tions can­cel each other out on the aggre­gate pri­vate bank­ing sec­tor reserve account, but increase the loan assets of the bank­ing sec­tor and the deposit lia­bil­i­ties at the same time. Reserves are thus needed to set­tle trans­ac­tions between indi­vid­ual banks, and as a con­duit for the gov­ern­ment sector’s finan­cial deal­ings with the pri­vate non-bank sec­tor, but at the aggre­gate level are nei­ther a require­ment for, nor a con­straint upon, pri­vate bank lend­ing.

Fig­ure 12: The Pri­vate Bank Sec­tor

The cru­cial flow equa­tions in this model are:

The change in the level of money in cir­cu­la­tion is thus deter­mined by net pri­vate sec­tor lend­ing, the gov­ern­ment deficit, and the mon­e­tary oper­a­tions of the Cen­tral Bank. This gen­er­al­izes, but does not con­tra­dict, the results of the ear­lier more sim­pli­fied model.


The endo­gene­ity of the money sup­ply means that banks, debt and money are essen­tial com­po­nents of macro­eco­nom­ics: they can­not be ignored as in stan­dard Neo­clas­si­cal macro­eco­nom­ics, nor car­i­ca­tured by “patient lends to impa­tient” bank-less mod­i­fied DSGE mod­els (Krug­man and Eggerts­son 2010). An alter­na­tive mon­e­tary macro­eco­nom­ics is required, and one can be con­structed eas­ily from the prin­ci­ples of dou­ble-entry book­keep­ing, as I have shown here.

Even at this rudi­men­tary level, this mon­e­tary per­spec­tive has lessons for the EU cri­sis. Since the pri­vate sec­tor is now delever­ag­ing, its actions are reduc­ing the money sup­ply. If the gov­ern­ment runs a sur­plus, this fur­ther reduces the money supply—and hence eco­nomic activ­ity. Aus­ter­ity there­fore ampli­fies the down­turn caused by pri­vate sec­tor delever­ag­ing.

At the same time, a gov­ern­ment deficit is less effec­tive than pro­po­nents of a gov­ern­ment stim­u­lus approach to the cri­sis have argued (Krug­man 2012f), since on the his­tor­i­cal evi­dence, pri­vate sec­tor delever­ag­ing is likely to con­tinue for a decade or more, given the level of accu­mu­lated pri­vate debt. Debt can­cel­la­tion will be required as well if pol­icy is to res­cue us from this cri­sis.


We present here a sim­ple way to rep­re­sent the con­cep­tual dif­fer­ence between spend­ing plans and cur­rent received income using the lan­guage of continuous–time func­tions with dis­con­ti­nu­ities.

Fig­ure 13: Change in debt as a jump func­tion

We start by defin­ing aggre­gate income flow as a func­tion with the prop­erty shown in Fig­ure 13, where rep­re­sents the value of the func­tion at time t and

cor­re­sponds to the limit of the val­ues of as time approaches t from the above (that is, from instants imme­di­ately after t, and there­fore after the injec­tion of new debt). The dif­fer­ence is a “jump” in the flow of aggre­gate income, which we argue is caused by a cor­re­spond­ing change in the stock of debt. Expressed in this way, equa­tion (1.8) sim­ply says that

That is, the flow of expen­di­ture always equals the ex-post flow of income, which in turn equals the flow of income plus change in the stock of debt. In par­tic­u­lar, observe that flows of expen­di­ture and income coin­cide except at the points of dis­con­ti­nu­ity, which are exactly the points when change in the stock of debt is assumed to occur.

Now the total expen­di­ture and income between times and is given by the inte­grals of the cor­re­spond­ing flows. But it fol­lows from the prop­er­ties of inte­gra­tion that func­tions dif­fer­ing only at finitely many points of dis­con­ti­nu­ities must have iden­ti­cal inte­grals. There­fore, pro­vided the change in debt occurs at finitely many points, we have that

which means that recorded expen­di­ture and income over a finite period , such as those found in NIPA tables, nec­es­sar­ily agree. This does not mean, how­ever, that changes in debt do not affect the inte­grals, but rather that they affect both inte­grals in exactly the same way. Put in another way, whereas com­par­ing recorded expen­di­ture and income over a period does not explic­itly reveal changes in debt, ignor­ing the effect of change in debt in the flows of expen­di­ture and income leads to a grossly incor­rect model.

Biggs, Michael and Mayer, Thomas (2010), ‘The Out­put Gap Conun­drum’, Intereconomics/Review of Euro­pean Eco­nomic Pol­icy, 45 (1), 11–16.

Biggs, Michael, Mayer, Thomas, and Pick, Andreas (2010), ‘Credit and Eco­nomic Recov­ery: Demys­ti­fy­ing Phoenix Mir­a­cles’, SSRN eLi­brary.

Blan­chard, Olivier (2009), ‘The State of Macro’, Annual Review of Eco­nom­ics, 1 (1), 209–28.

Blan­chard, Olivier J. (2008), ‘The State of Macro’, SSRN eLi­brary.

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

deLong, J. Brad­ford (2012), ‘The Per­ils of Prophecy’, Project Syn­di­cate (New York: Project Syn­di­cate).

ECB (2012), ‘Mon­e­tary and Finan­cial Devel­op­ments’, Monthly Bul­letin May 2012 (Brus­sels: Euro­pean Cen­tral Bank).

Fama, Eugene F. and French, Ken­neth R. (1999), ‘The Cor­po­rate Cost of Cap­i­tal and the Return on Cor­po­rate Invest­ment’, Jour­nal of Finance, 54 (6), 1939–67.

Fiebiger, Brett (2012), ‘Mod­ern Money The­ory And The ‘Real-World’ Account­ing Of 1–1<0: The U.S. Trea­sury Does Not Spend As Per A Bank’, PERI Polit­i­cal Econ­omy Research Insti­tute (Amherst: Uni­ver­sity of Mass­a­chu­setts), 1–16.

Fried­man, Mil­ton (1969), ‘The Opti­mum Quan­tity of Money’, The Opti­mum Quan­tity of Money and Other Essays (Chicago: MacMil­lan), 1–50.

— (1997), ‘The Euro: Mon­e­tary Unity To Polit­i­cal Dis­unity?’, Project Syn­di­cate, 1997 (August 28).

Full­wiler, Scott T., Kel­ton, Stephanie, and Wray, Ran­dall (2012), ‘Mod­ern Money The­ory: A Response To Crit­ics’, PERI Polit­i­cal Econ­omy Research Insti­tute (Amherst: Uni­ver­sity of Mass­a­chu­setts), 17–26.

God­ley, Wynne (1992), ‘Maas­tricht and All That’, Lon­don Review of Books, 14 (19), 3–4.

Hicks, J. R. (1935), ‘Wages and Inter­est: The Dynamic Prob­lem’, The Eco­nomic Jour­nal, 45 (179), 456–68.

— (1937), ‘Mr. Keynes and the “Clas­sics”; A Sug­gested Inter­pre­ta­tion’, Econo­met­rica, 5 (2), 147–59.

Hicks, John (1981), ‘IS-LM: An Expla­na­tion’, Jour­nal of Post Key­ne­sian Eco­nom­ics, 3 (2), 139–54.

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

Keen, Steve ‘A mon­e­tary Min­sky model of the Great Mod­er­a­tion and the Great Reces­sion’, Jour­nal of Eco­nomic Behav­ior & Orga­ni­za­tion, (0).

— (1995), ‘Finance and Eco­nomic Break­down: Mod­el­ing Minsky’s ‘Finan­cial Insta­bil­ity Hypoth­e­sis.”, Jour­nal of Post Key­ne­sian Eco­nom­ics, 17 (4), 607–35.

— (2012), ‘Krug­man Apol­o­gises!’, in Steve Keen (ed.), Steve Keen’s Debt­watch (Syd­ney: Cen­tre for Eco­nomic Sta­bil­ity Incor­po­rated).

Keynes, J. M. (1936), The gen­eral the­ory of employ­ment, inter­est and money (Lon­don: Macmil­lan).

— (1937), ‘Alter­na­tive the­o­ries of the rate of inter­est’, Eco­nomic Jour­nal, 47, 241–52.

Krug­man, Paul (2009), ‘Actu­ally exist­ing Min­sky’, The Con­science of a Lib­eral (New York: New York Times).

— (2012a), ‘A Teach­able Money Moment’, The Con­science of a Lib­eral (New York: New York Times).

— (2012b), ‘Things I Should Not Be Wast­ing Time On’, The Con­science of a Lib­eral (New York: New York Times).

— (2012c), ‘Min­sky and Method­ol­ogy (Wonk­ish)’, The Con­science of a Lib­eral (New York: New York Times).

— (2012d), ‘Bank­ing Mys­ti­cism, Con­tin­ued’, The Con­science of a Lib­eral (New York: New York Times).

— (2012e), ‘Delever­ag­ing, Mon­e­tary Pol­icy, and Fis­cal Pol­icy: A Fur­ther Note’, The Con­science of a Lib­eral (New York: New York Times).

— (2012f), End this Depres­sion Now! (New York: W.W. Nor­ton).

— (2012g), ‘Oh My, Steve Keen Edi­tion’, The Con­science of a Lib­eral (New York: New York Times).

Krug­man, Paul and Eggerts­son, Gauti B. (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, Finn E. and Prescott, Edward C. (1990), ‘Busi­ness Cycles: Real Facts and a Mon­e­tary Myth’, Fed­eral Reserve Bank of Min­neapo­lis Quar­terly Review, 14 (2), 3–18.

Min­sky, Hyman P. (1982), Can “it” hap­pen again? : essays on insta­bil­ity and finance (Armonk, N.Y.: M.E. Sharpe) xxiv, 301 p.

— (1986), Sta­bi­liz­ing an unsta­ble econ­omy (Twen­ti­eth Cen­tury Fund Report series, New Haven and Lon­don: Yale Uni­ver­sity Press) xiv.

Moore, Basil J. (1979), ‘The Endoge­nous Money Stock’, Jour­nal of Post Key­ne­sian Eco­nom­ics, 2 (1), 49–70.

— (1983), ‘Unpack­ing the Post Key­ne­sian Black Box: Bank Lend­ing and the Money Sup­ply’, Jour­nal of Post Key­ne­sian Eco­nom­ics, 5 (4), 537–56.

— (1988), Hor­i­zon­tal­ists and Ver­ti­cal­ists: The Macro­eco­nom­ics of Credit Money (Cam­bridge: Cam­bridge Uni­ver­sity Press).

— (1997), ‘Rec­on­cil­i­a­tion of the Sup­ply and Demand for Endoge­nous Money’, Jour­nal of Post Key­ne­sian Eco­nom­ics, 19 (3), 423–28.

O’Brien, Yueh-Yun June C. (2007), ‘Reserve Require­ment Sys­tems in OECD Coun­tries’, SSRN eLi­brary.

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

Wray, L. R. (2011), ‘MMT: A Dou­bly Ret­ro­spec­tive Analy­sis’, New Eco­nomic Per­spec­tives.

Wray, Larry Ran­dall (1988), ‘Profit Expec­ta­tions and the Invest­ment-Sav­ing Rela­tion’, Jour­nal of Post Key­ne­sian Eco­nom­ics, 11 (1), 131–47.



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.
  • Hi Michael,

    It was lit­er­ally the first pass, so there’s no pub­lished paper as yet.

    I’ll check those entries when I have time–and write a paper based on it–but I think they are all OK. I went over them–if quickly–with Scott, Stephanie and Matheus, and none of them spot­ted any erro­neous signs.

  • Michael Bridges

    Thanks Steve. I can wait on the paper but I do want to start think­ing about what you have released so far. Am I cor­rect that the vari­ables Gov, Tax and sellBD are flows? Fur­ther­more is it true that Gov and Tax are always non-neg­a­tive vari­ables such that the trea­sury Deposit account should go down if Gov > Tax (i.e. spend­ing exceeds income) and no money is bor­rowed i.e. sellBD=0?

    Finally it appears that the for any deficit (i.e. Gov > Tax) there must be an equiv­a­lent amount of money bor­rowed (i.e. sellBD=Gov-Tax). Does this mean that GD, the trea­sury deposit account held at the CB, remains con­stant i.e. d/dt(GD)=0?

    Can any­one else out there com­ment?


  • Michael Bridges

    With regards to d/dt(GD)=0, upon fur­ther think­ing, this con­di­tion can occur when Gov > Tax but is not nec­es­sar­ily the only con­di­tion. I see that sellBD can be greater than or equal to the deficit i.e. sellBD>=Gov-Tax. In other words the Gov­ern­ment can sell more bonds than it needs to cover the deficit in which case the GD can increase. Does this occur in real­ity?

  • Pingback: European disunion and endogenous money « Real-World Economics Review Blog()

  • Alex Coven­try

    I know it’s a minor part of the paper, but I think the whole part about propo­si­tion 1 in the “Test­ing the ‘Liq­uid­ity Trap’” sec­tion could use a rewrite, as I found it quite con­fus­ing. To begin with, it’s not clear to me how that fol­lows from Krugman’s argu­ment, though per­haps only because I am not an econ­o­mist.

  • Alex Coven­try

    Read­ing on, fig­ure 7 doesn’t clar­ify any­thing at all for me. Per­haps it would if I knew the dou­ble count­ing argu­ment you are rebut­ting.

  • Alex Sefer­ian

    Dear Pro­fes­sor Steve,
    In addi­tion to this post, I reviewed the Fields Sem­i­nar videos. I agree with, and applaud your efforts in rela­tion to, the need to develop a robust alter­na­tive to main­stream macro­eco­nom­ics. To the extent that both MMT and MCT have recently ben­e­fited from increased expo­sure in the media, and to the extent that both plat­forms ben­e­fit from impor­tant ide­o­log­i­cal com­mon­al­i­ties, uni­fy­ing them would be a great step for­ward indeed.
    I also wanted to say that your math­e­mat­i­cal work is very valu­able to peo­ple like myself who find the process of accept­ing a new the­o­ret­i­cal frame­work eas­ier when con­cepts can be coher­ently mod­elled with for­mu­las. In that respect, I very much look for­ward to a future post that you men­tion and that will shed fur­ther light on the dou­ble-count­ing debate, as well as address con­cerns such as those posted by JKH on July 3.
    At the same time, I sym­pa­thize with some who have com­mented on the dif­fi­cul­ties of grasp­ing parts of your work. For those peo­ple who may be per­haps less num­bers-ori­ented, all I would say is that the maths may be con­sid­ered a “nec­es­sary evil”. How­ever, main­stream macro­eco­nom­ics will only be truly debunked when sim­ple con­cepts are under­stood by non-econ­o­mists and informed com­men­ta­tors – with­out them nec­es­sar­ily hav­ing to ref­er­ence the for­mu­las and the num­bers to reach their own con­clu­sions.
    Bear­ing the above in mind, I was hop­ing you could com­ment on what your lat­est feed­back may be on the chances of MMT and MCT ulti­mately con­verg­ing? In terms of the maths, your approach comes across to me as MMT “taken to the next level” (as long the dou­ble count­ing debate is firmly closed). Is that a fair state­ment, or are there fun­da­men­tal dif­fer­ences that were unearthed dur­ing the Fields Sem­i­nar? More than just a ques­tion of what for­mu­las to ref­er­ence, it seems to me that dif­fer­ent pol­icy rec­om­men­da­tions may result depend­ing on what approach one adopts. Wayne Godley’s sec­to­r­ial bal­ances approach sug­gests that gov­ern­ment deficits are needed for the pri­vate sec­tor to de-lever, if the cur­rent account is in bal­ance. Your approach sug­gests that a gov­ern­ment deficit may not be enough, as this may sim­ply lead to the pri­vate sec­tor increas­ing its spec­u­la­tive asset acqui­si­tions.
    More broadly, do MMT and MCT, in your opin­ion, advo­cate dif­fer­ent solu­tions to address the euro cri­sis? From what I have picked up, the answer — before the Fields Sem­i­nar — was yes. Is that still the case?
    Regard­ing Europe, MMT’s pol­icy stance includes the fol­low­ing argu­ments: i) gov­ern­ment deficits are key as the region’s prob­lems relate mainly to a lack of aggre­gate demand, ii) the pri­vate sec­tor can­not ade­quately gen­er­ate enough demand until it repairs its bal­ance sheet, iii) infla­tion and employ­ment are among the most impor­tant vari­ables to mon­i­tor — deficits should be cal­i­brated so as to cre­ate full employ­ment, but not infla­tion (gov­ern­ment debt-to-GDP ratios are close to mean­ing­less).
    It would seem to me – again, before the Fields Sem­i­nar – that your pol­icy stance regard­ing Europe is quite dif­fer­ent. I under­stand that you argue that: i) a national gov­ern­ment can­not over­come the real down­turn in the pri­vate sec­tor by financ­ing its bud­get deficit via debt, ii) the cri­sis was caused by too much debt, and get­ting into more debt is not going to solve the prob­lem, and iii) the solu­tion to the debt prob­lem is for gov­ern­ments to get rid of the debt by either: a) financ­ing deficits via the print­ing of money, thereby caus­ing infla­tion, and in turn erod­ing the value of the debt, or b) sim­ply can­celling the debt out­right (a debt jubilee). MMTers as far as I have been able to read never men­tion a debt jubilee, and would argue that print­ing money is nei­ther nec­es­sar­ily infla­tion­ary nor nec­es­sary to finance a deficit.
    Apolo­gies in advance to you and/or to MMTers if I have mis­rep­re­sented the pol­icy stances. I look for­ward to fol­low-ups to this post and in the spirit of fos­ter­ing a healthy debate to fur­ther the goal of cre­at­ing an alter­na­tive to main­stream macro­eco­nom­ics.

  • Bhaskara II

    Inter­est­ing for­mula.

    Rear­rang­ing the terms in a for­mula above to show income, Y_I, gives income reduced by change in debt or change in debt reduces the income, or a com­bi­na­tion. Y_I being some what aggre­gated.

  • Bhaskara II

    Inter­est­ing for­mula.

    Rear­rang­ing the terms in a for­mula above to show income, Y_I, gives income reduced by change in debt, change in debt reduces the income, or a com­bi­na­tion. Y_I being some what aggre­gated (with­out the banks).

  • Bhaskara II

    Or, the for­mula shows debt as con­tra-sav­ings.

  • Bhaskara II

    Pro­fes­sor Keen and oth­ers might be inter­ested in this:

    Here is a empir­i­cal paper on busi­ness cycle and finan­cial cycle fre­quen­cies that might be of inter­est. They sep­a­rate a higher fre­quency busi­ness cycle com­pared to a rel­a­tively lower fre­quency finan­cial cycle. In the begin­ning they seem to be in agree­ment with a few of Prof. Keen’s argu­ments.

    It cov­ers mul­ti­ple economies and time peri­ods. I have only skimmed a few parts and quickly skimmed an other paper on the band pass fil­ter that they use. They may not have used long enough data series to eval­u­ate longer cycles.

    “We char­ac­terise empir­i­cally the finan­cial cycle using two approaches: analy­sis of turn­ing points and fre­quency-based fil­ters. We iden­tify the finan­cial cycle with the medium-term com­po­nent in the joint fluc­tu­a­tions of credit and prop­erty prices; equity prices do not fit this pic­ture well. We show that finan­cial cycle peaks are very closely asso­ci­ated with finan­cial crises and that the length and ampli­tude of the finan­cial cycle have increased markedly since the mid-1980s. We argue that this reflects, in par­tic­u­lar, finan­cial lib­er­al­i­sa­tion and changes in mon­e­tary pol­icy frame­works. So defined, the finan­cial cycle is much longer than the tra­di­tional busi­ness cycle. Busi­ness cycle reces­sions are much deeper when they coin­cide with the con­trac­tion phase of the finan­cial cycle. We also draw atten­tion to the “unfin­ished reces­sion” phe­nom­e­non: pol­icy responses that fail to take into account the length of the finan­cial cycle may help con­tain reces­sions in the short run but at the expense of larger reces­sions down the road.”

    They agree with Prof. Keen in the intro­duc­tion:
    “The dom­i­nant pre-cri­sis par­a­digms viewed finance largely as a sideshow to macro­eco­nomic fluctuations.[2] The cri­sis demon­strated that this pre­sump­tion was dan­ger­ously wrong. In con­trast to what much of the lit­er­a­ture assumed, inter­est rates could not cap­ture all the inter­ac­tions between the finan­cial and real sides of the econ­omy. And even strands of work that incor­po­rated richer ele­ments of those inter­ac­tions, such as the role of col­lat­eral (eg, Bernanke et al (1999)), fell way short of repli­cat­ing the strength and occa­sional vir­u­lence of the processes at work. After all, they could not account for finan­cial crises. A rapidly grow­ing lit­er­a­ture is now seek­ing to rem­edy these short­com­ings. The pre­vail­ing approach is to incor­po­rate richer finan­cial sec­tors into dynamic sto­chas­tic gen­eral equi­lib­rium (DSGE) models.[3] Whether this line of enquiry will ulti­mately prove fruit­ful is a legit­i­mate sub­ject of debate.[4]”

    Regard­less of the spe­cific approach, how­ever, any future work to model finan­cial fac­tors requires a bet­ter under­stand­ing of the stylised empir­i­cal reg­u­lar­i­ties of the “finan­cial cycle”, with its booms and busts pos­si­bly lead­ing to seri­ous finan­cial and macro­eco­nomic strains. And yet, the mean­ing and char­ac­ter­i­sa­tion of the finan­cial cycle remain elu­sive.”

    Against this back­drop, the objec­tive of our paper is to con­tribute to a bud­ding lit­er­a­ture that seeks to char­ac­terise the finan­cial cycle. Some of this work has shed only indi­rect light on the ques­tion. It has done so in var­i­ous ways: by doc­u­ment­ing empir­i­cally the behav­iour of the rela­tion­ship between credit, asset prices and real eco­nomic activ­ity (eg, Borio et al (1994), Detken and Smets (2004), Good­hart and Hoff­man (2008), Schu­lar­ick and Tay­lor (2009)); by devel­op­ing lead­ing indi­ca­tors of finan­cial dis­tress (eg, Borio and Lowe (2002), Alessi and Detken (2009), Gerdesmeier et al (2010)); and by exam­in­ing the fore­cast­ing prop­er­ties for eco­nomic activ­ity of var­i­ous finan­cial indi­ca­tors beyond inter­est rates (eg, Eng­lish et al (2005), Borio and Lowe (2004), Ng (2011), Hatz­ius et al (2010)).”

    They look at a credit to GDP ratio. (Credit / debt.)
    “In these cases, pol­i­cy­mak­ers may focus too much on equity prices and stan­dard busi­ness cycle mea­sures and lose sight of the con­tin­ued build-up of the finan­cial cycle, char­ac­terised by fur­ther increases in the ratio of credit to GDP and in prop­erty prices. The bust that fol­lows an unchecked finan­cial boom brings about large eco­nomic dis­rup­tions.”


    Cycles Paper:
    Math­ias Drehmann, Clau­dio Borio and Kostas Tsat­sa­ro­nis, “Char­ac­ter­is­ing the finan­cial cycle: don’t lose sight of the medium term!”, BIS Work­ing Papers No. 380; June 2012.

    Band Pass Fil­ter:

  • Pingback: Clint Ballinger - Modern Monetary Theory & Full Reserve Banking: Connected by Fiat()