Research

Flattr this!

This page lists and links to my aca­d­e­mic papers. If you would like to sup­port my research, please click on this link: Research Fund­ing.

Research Financial Instability and Endogenous Money.

A mon­e­tary Min­sky model of the Great Mod­er­a­tion & the Great Reces­sion. This paper is forth­com­ing in the Jour­nal of Eco­nomic Behav­ior and Orga­ni­za­tion. The sim­u­la­tions there are  slightly mis­lead­ing since the ini­tial con­di­tions con­tained an incon­sis­tency; these have since been cor­rected to yield the same long term out­come but far less volatile ini­tial fluc­tu­a­tions. I’ll pub­lish the paper with these revised con­di­tions shortly

A model of endoge­nous credit cre­ation and a credit crunch. This paper is very tech­ni­cal and out­lines my analy­sis of a credit crunch, which shows that if that was the only prob­lem we faced, a gov­ern­ment res­cue could work, and con­trary to stan­dard mon­e­tary the­ory (a.k.a. the “money mul­ti­plier” model) it would be much bet­ter to give the gov­ern­ment money to debtors than to the banks. It also out­lines my pre­lim­i­nary mul­ti­sec­toral mon­e­tary model of pro­duc­tion. This paper was pro­duced with the finan­cial assis­tance of theP­aul Wool­ley Cen­tre for Cap­i­tal Mar­ket Dys­func­tion­al­ity at the Uni­ver­sity of Tech­nol­ogy, Syd­ney.

House­hold Debt—the final stage in an arti­fi­cially extended Ponzi Bub­ble, Aus­tralian Eco­nomic Review, Vol. 42 No. 3 Sep­tem­ber 2009, pp. 347–57. This paper shows that the growth in house­hold debt was not an equi­lib­rium response to falling inter­est rates, but a Ponzi spec­u­la­tive bub­ble whose burst­ing is caus­ing a seri­ous reces­sion. I present a model of Minsky’s “Finan­cial Insta­bil­ity Hypoth­e­sis” that includes Ponzi finance as well as pro­duc­tive invest­ment; the model gen­er­ates a Depres­sion when debt accu­mu­lated for spec­u­la­tive pur­poses over­whelms the pro­duc­tive capac­ity of the econ­omy.

The Dynam­ics of the Mon­e­tary Cir­cuit, in The Polit­i­cal Econ­omy Of Mon­e­tary Cir­cuits: Tra­di­tion And Change In Post-Key­ne­sian Eco­nom­ics, edited by Jean-François Pon­sot and Ser­gio Rossi (Pal­grave, 2009, pp. 161–187). This is a rea­son­ably acces­si­ble expla­na­tion of the tech­nique I use to derive dynamic mod­els of finance, and advo­cacy of con­tin­u­ous time meth­ods over the dis­crete time approach that dom­i­nates Post Key­ne­sian eco­nom­ics today.

Bail­ing out the Titanic with a Thim­bleEco­nomic Analy­sis and Pol­icy, Vol 39 Issue 1, pp. 3–24. Unlike most ref­er­eed aca­d­e­mic jour­nals, this one is freely acces­si­ble online. In this paper I explain why I don’t expect the bailouts to work, I present a model of a credit crunch in a pure credit econ­omy where the credit crunch alone causes a Great Depres­sion.

My PhD the­sis Eco­nomic Growth and Finan­cial Insta­bil­ity (UNSW, 1997).

Using Math­cad in Eco­nomic Analy­sis

This arti­cle, on Hearne Sci­en­tific Software’s web­site, explains my use of Math­cad for both data analy­sis and mod­el­ling.

The Non­lin­ear Dynam­ics of Debt Defla­tion. This tech­ni­cal paper gives a math­e­mat­i­cal model of Minsky’s Finan­cial Insta­bil­ity Hypoth­e­sis. It includes an extended model with a gov­ern­ment sec­tor that can con­tain the process of pri­vate debt accu­mu­la­tion, and a pre­lim­i­nary attempt to model price dynam­ics

The Non­lin­ear Eco­nom­ics of Debt Defla­tion. This tech­ni­cal paper also has a gov­ern­ment sec­tor exten­sion to the basic non-mon­e­tary Min­sky model from my 1995 paper, and shows that under some cir­cum­stances there is a bifur­ca­tion in gov­ern­ment debt when sta­bi­liz­ing an unsta­ble econ­omy: in some cir­cum­stances, both pri­vate debt and gov­ern­ment debt sta­bi­lize as a per­cent­age of GDP, but in oth­ers run­away gov­ern­ment debt is needed to sta­bi­lize pri­vate debt.

The non-con­ser­va­tion of money

The process of endoge­nous money cre­ation

The “Finan­cial Insta­bil­ity Hypoth­e­sis”

Expert Opin­ion on Ponzi Loans

Other Topics

I’ll grad­u­ally link all my aca­d­e­mic papers here, since jour­nals now seem com­fort­able about aca­d­e­mics link­ing their papers on their own web­sites.

Econo­physics

Wor­ry­ing trends in econo­physics. Mauro Gal­le­gati, Steve Keen, Thomas Lux, Paul Ormerod, (2006). “Wor­ry­ing trends in econo­physics”, Phys­ica A 370, pp. 1–6.

Marx

A Marx for Post Key­ne­sians; unpub­lished

The Mis­in­ter­pre­ta­tion of Marx’s The­ory of Value; Jour­nal of the His­tory of Eco­nomic Thought, 15 (2), Fall, 282–300

Use-value, exchange-value, and the demise of Marx’s Labour The­ory of Value; Jour­nal of the His­tory of Eco­nomic Thought, 15 (1), Spring, 107–121

Use, Value and Exchange: The Mis­in­ter­pre­ta­tion of Marx; my Mas­ters the­sis on Marx, UNSW 1990; unpub­lished

Theory of the Firm

The con­ven­tional the­ory of com­pe­ti­tion is non­sense. I explain why in Chap­ter 4 of Debunk­ing Eco­nom­ics, “Size Does Mat­ter”. A more tech­ni­cal expla­na­tion is given in this paper:

Steve Keen (2004). “Dereg­u­la­tor: Judg­ment Day for Micro­eco­nom­ics”, Util­i­ties Pol­icy, 12: 109 –125

This only cov­ers the Mar­shal­lian the­ory how­ever. More detailed cri­tiques that are rel­e­vant to the Cournot model as well are pub­lished here:

Steve Keen and Rus­sell Stan­dish, (2006). “Profit Max­i­miza­tion, Indus­try Struc­ture, and Com­pe­ti­tion: A cri­tique of neo­clas­si­cal the­ory”, Phys­ica A 370: 81–85

Steve Keen and Rus­sell Stan­dish, (2010). “Debunk­ing the the­ory of the firm—a chronol­ogy”, real-world eco­nom­ics review, issue no. 53, 26 June 2010, pp. 56–94, http://www.paecon.net/PAEReview/issue53/KeenStandish53.pdf

This paper gives a com­plete chrono­log­i­cally laid out cov­er­age of our cri­tique, from its begin­nings when writ­ing Debunk­ing Eco­nom­ics to a demon­stra­tion that the Cournot-Nash equi­lib­rium is meta-unsta­ble.

Say’s Law

Nudge Nudge, Wink Wink Say No More!” in Steve Kates (ed.), Two Hun­dred Years of Say’s Law, Edward Elgar, Alder­shot, pp. 199–209.

Transna­tional cor­po­ra­tions and aggre­gate demand

The relo­ca­tion of pro­duc­tion and aggre­gate demand (Tech­ni­cal Appen­dix)

  • John Hawkins

    I think accord­ing to your def­i­n­i­tion that this would be the “veloc­ity” of money as Richard is think­ing of it.

    http://research.stlouisfed.org/fred2/graph/fredgraph.pdf?&chart_type=line&graph_id=&category_id=&recession_bars=On&width=630&height=378&bgcolor=%23b3cde7&graph_bgcolor=%23ffffff&txtcolor=%23000000&ts=8&preserve_ratio=true&fo=ve&id=GDP_BASE_TCMDO&transformation=lin_lin_lin&scale=Left&range=Max&cosd=1947–01-01&coed=2012–08-22&line_color=%230000ff&link_values=&mark_type=NONE&mw=4&line_style=Solid&lw=1&vintage_date=2012–08-29_2012-08-29_2012-08–29&revision_date=2012–08-29_2012-08-29_2012-08–29&mma=0&nd=__&ost=&oet=&fml=a%2F%28b%2Bc%29&fq=Quarterly&fam=avg&fgst=lin

    or it’s pos­si­ble he may be think­ing more along the lines of the Mar­ginal Prod­uct of Debt which has been made famous in many “debt sat­u­ra­tion” pre­sen­ta­tions (change in gdp / change in debt)

    http://research.stlouisfed.org/fred2/graph/fredgraph.pdf?&chart_type=line&graph_id=&category_id=&recession_bars=On&width=630&height=378&bgcolor=%23b3cde7&graph_bgcolor=%23ffffff&txtcolor=%23000000&ts=8&preserve_ratio=true&fo=ve&id=GDP_TCMDO&transformation=ch1_ch1&scale=Left&range=Max&cosd=1947–01-01&coed=2012–04-01&line_color=%230000ff&link_values=&mark_type=NONE&mw=4&line_style=Solid&lw=1&vintage_date=2012–08-29_2012-08–29&revision_date=2012–08-29_2012-08–29&mma=0&nd=_&ost=&oet=&fml=a%2Fb&fq=Quarterly&fam=avg&fgst=lin

    Either way, they are hardly con­stant so it would be hard to plug it into a for­mula

  • Robin North­cott

    Hi Steve,

    I’m con­fused about the sec­ond model in the first paper here (the extended God­ley model includ­ing debt). It looks to me like there is an incon­sis­tency.

    Eqn 1.4 (rearranged) gives Y = Pi + w.L + r.D
    but Y must also be Y = w.L + I (every­thing is bought either for con­sump­tion or invest­ment).
    Sub­sti­tut­ing Eqn 1.3 in this eqn gives Y = w.L + Pi + dD/dt
    So dD/dt = r.D which seems a strange restric­tion and not actu­ally in the model so far as I can tell.

    I’m prob­a­bly just miss­ing some­thing here but an expla­na­tion would be great.

    Thanks

    Robin

  • Robin North­cott

    Think­ing about this a bit more I think I may under­stand what’s going on.
    Work­ers must be sav­ing (to lend to firms) so the con­sump­tion equa­tion becomes (adding inter­est pay­ments to work­ers income)
    Y = (1-S).(w.L+r.D) + I = (1-S).(w.L+r.D) + Pi + dD/dt
    =Pi + w.L + r.D + dD/dt — S.(w.L + r.D)
    so look­ing at eqn 1.4 again we get:
    dD/dt = S.(w.L + r.D) (which now seems pretty obvi­ous)

  • Dear Steve:

    I have been work­ing recently on an extended ver­sion of the Good­win model using meth­ods sim­i­lar to your own. Refer­ring to your paper from 1995 on your mod­i­fied Good­win model (pp.615–617), where invest­ment is a func­tion of profit rate, and there is no House­hold sav­ings (all worker income goes to con­sump­tion), I have the fol­low­ing com­ment:
    Unless I have mis­un­der­stood some­thing, it would seem that your model vio­lates the sec­tor bal­ance equa­tion, which is in this case:
    Y-WL-rD –I (pro­duc­tion sec­tor sur­plus)+ 0 (House­hold sec­tor sur­plus) + rD (Bank sec­tor sur­plus) =0.

    Thus I= Y-WL, or I=Y(1-W/a). Any other value of I would require House­hold sav­ings to be non-zero, vio­lat­ing the assump­tion of no House­hold sav­ings.
    I would very much like to hear your com­ment on this, as my model assumes that the sec­tor bal­ance equa­tion above must be sat­is­fied at all times, which gives a very dif­fer­ent result. Have I mis­un­der­stood some­thing?
    Look­ing for­ward to hear­ing from you.

    Best regards,
    Rojo­nia

  • Hi Rojo­nia,

    The model assumes that all non-invest­ment out­put is con­sumed: it deter­mines Y and I but leaves C unspec­i­fied. It is stock-flow con­sis­tent given that. Check for papers by Matheus Gras­selli and Bernardo Costa-Lima on this. Their research group is also work­ing on intro­duc­ing stocks so that both con­sump­tion and invest­ment can be deter­mined in a fully spec­i­fied model.

  • Herib­er­toAr­ribas

    Mr. Keen, a lit­tle joke:

    If Q is the suma­tory from i = 1 to n the q_i and Q is a con­stant when n increase indef­i­nitely q_i become zero.

    Then each firm fac­ing a won­der­full hor­i­zon­tal demand POINT q_i = 0

  • Nice one, thanks.

  • Tim­o­thée Thierry

    Hello,

    First of all I love your work ! I was one of the dis­ap­pointed stu­dent in neo­clas­si­cal eco­nom­ics that you’ve described so well in “Debunk­ing eco­nom­ics”. I decided to study math­e­mat­ics after my eco­nom­ics mas­ter degree but I was still very inter­ested in eco­nom­ics. When I read your book I was really happy to dis­cover that other sig­nif­i­cant the­o­ries already existed, since at this time I thought almost every­thing had to be done, which was quite dis­cour­ag­ing.

    I’m try­ing to fully under­stand your mon­e­tary model of a cap­i­tal­ist econ­omy and I think I get most of it but there is one ratio­nale, that seems cru­cial to me, that still resists me. I can’t really grasp what are the mean­ing and the val­ues you give at the para­me­ters beta and omega at page 3 of the arti­cle “Solv­ing the Para­dox of Mon­e­tary Prof­its”.

    You say first that “E and F will be some fac­tor (say omega and beta respec­tively) of the bal­ances in the accounts WD and BI; thus E=omega WD and F=beta BI.” where E and F are flows from the Worker Deposit account and from the Bank Income account to the Firm Deposit account.

    Then you say that “[omega and beta] describe the rate of flow of money out of the BI and WD accounts respec­tively to pay for con­sump­tion.” If I refer to this, I would under­stand that beta = con­sump­tion of banks per unit of time / aver­age amount of money in Bl per unit of time, and sim­i­larly for omega. It means I would inter­pret it as the aver­age propen­sion to con­sume of bankers and work­ers.

    But the next sen­tence asserts that “The val­ues of omega and beta are there­fore numer­i­cal esti­mates of how often bankers and work­ers respec­tively turn over the bal­ances in their accounts in a year.[…] These para­me­ters are thus the ratios between con­sump­tion over a year, and the aver­age bal­ance in the respec­tive accounts at any point in the year. A rea­son­able value for omega would be 26, since work­ers tend to spend their wages on a fort­nightly basis.” This part is very con­fus­ing for me. As I under­stand it, it would mean that if I have an aver­age of 1000€ on my bank account dur­ing the year (mean­ing per­haps I earn 2000€ per month), I would con­sume 26000€ over the year when earn­ing 24000€. So the only thing I’m cer­tain is that it is not the good inter­pre­ta­tion ! I don’t really get how one can derived this ratio of 26 for “a rea­son­able value” of omega and how to inter­pret it. So maybe you can help me with this since I think it is one cru­cial ratio­nale if I want to explain to some peo­ple the other mod­eli­sa­tion of econ­omy one can do !

    I don’t know if you’re still active on this blog but I hope you can give me some clues ! I’ll keep search­ing in the mean­time.

    Best regards,
    Tim­o­thée Thierry