SASE 2011 Pre­sen­ta­tion: The Fail­ure of Neo­clas­si­cal Macro & the Mon­e­tary Cir­cuit The­ory Alter­na­tive

Flattr this!

This is the pre­sen­ta­tion I gave at the 2011 con­fer­ence of the Soci­ety for the Advance­ment of Socio-Eco­nom­ics (SASE) annual con­fer­ence in Madrid last week:

It com­bines four themes that will be promi­nent in my pub­lic talks from now on:

  • Neo­clas­si­cal econ­o­mists don’t under­stand neo­clas­si­cal eco­nom­ics;
  • Neo­clas­si­cal “rep­re­sen­ta­tive agent” macro­eco­nom­ics (both the so-called New Clas­si­cal and New Key­ne­sian vari­ants) vio­late fun­da­men­tal research by neo­clas­si­cal econ­o­mists into the foun­da­tions of neo­clas­si­cal the­ory;
  • The Credit Accel­er­a­tor explains the Great Depres­sion and the Great Reces­sion (here my argu­ments are sim­i­lar to those of Richard Koo, and the Credit Accel­er­a­tor is the same con­cept ini­tially derived by Biggs, Mayer and Pick that they called the Credit Impulse); and
  • The “Mon­e­tary Cir­cuit The­ory” model (my thanks to Mike Hon­ey­church for sug­gest­ing this name for my approach).

I had some soft­ware hassles–I changed the screen res­o­lu­tion after load­ing my screen cap­ture pro­gram, which caused it to crash!–so I had to “dub” the talk later, and so there’s a bit of chat­ter about that at the start, and the sound lev­els are rather low. So my apolo­gies, but it’s all I had time for when also cop­ing with a brand new HP lap­top that has been hav­ing BSOD events so reg­u­larly that I am now pin­ing for Mephistophe­les, my Dell lap­top that I left back in Syd­ney.

In the 30 min­utes I had, I only cov­ered about half the mate­r­ial in the Pow­er­point slides; a sim­i­lar issue will apply when I present at the Cen­tral Bank of Argentina’s annual con­fer­ence on Thurs­day. Hope­fully one day I’ll get the chance to present this talk over an hour or so, which would be needed to cover all the infor­ma­tion in it.

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.
  • TruthIs­ThereIs­NoTruth

    hello Steve,

    Could you please explain slides 44 and 45. This is really con­fus­ing me, my under­stand­ing is that if you lag a vari­able, you are bring­ing past val­ues for­ward, there­fore the lagged vari­able is the explan­o­tary vari­able. I am at pain to under­stand where my under­stand­ing is incor­rect in this case.

  • RickW

    How widely accepted is your mod­el­ling? Who is lis­ten­ing — apart from me of course! 

    Have you tested what sub­tle changes can be made to resta­bilise the sys­tem once it is headed for col­lapse? Were you seri­ous sug­gest­ing that there should be wage increases?

  • Hi TININT,

    What I mean is that changes in the CA pre­cede changes in stock and prop­erty prices. But as you know, my spe­cialty is math­e­mat­i­cal mod­el­ing rather than stats (let alone eCONo­met­rics), so my lan­guage here can be sloppy, and I have used rou­tines I’ve writ­ten in Math­cad to do the stats there rather than a stan­dard econo­met­ric pack­age.

    Given com­ments here that the tim­ing looks the other way around on the images, I’ve sent the data to a col­league to check with stan­dard sta­tis­ti­cal pack­ages, in case I did make a mis­take in my rou­tines. Once I get word back–one way or the other–from that col­league, I’ll post the results and the orig­i­nal data here.

  • Accep­tance is pretty wide in the finance com­mu­nity these days Rick, and I am in Argentina right now as a guest of their Cen­tral Bank. Within the Post Key­ne­sian camp there are disputes–the Char­tal­ist-Cir­cuitist issue–but my work is influ­en­tial there as well, if decid­edly unpop­u­lar with some indi­vid­u­als in that camp.

    On the wages issue, infla­tion would reduce the bur­den of out­stand­ing pri­vate sec­tor debt, and a wage rise would be a much more direct way to cause infla­tion than fid­dling with Base Money, which is the cur­rent US Fed pol­icy (com­bined with defla­tion­ary bud­get deficit reduc­tions by the Fed­eral Gov­ern­ment). But there’s Buckley’s chance of that pol­icy being fol­lowed in prac­tice of course.

    And test­ing these mod­els is a long way down the track. If I had a research team, or if there was an entire com­mu­nity of schol­ars (as the neo­clas­si­cals have with their mis­guided the­o­ries), then it would have already hap­pened. But as a one man show, it sim­ply hasn’t been pos­si­ble for me to go beyond build­ing them in the first instance.

  • Neb­bi­olo71

    Hi Steve,

    I’m con­tin­u­ously fol­low­ing your work since more than a year now but I still haven’t really fig­ured out exactly what are the dis­pute between you and the char­tal­ists (I sup­pose espe­cially Bill Mitchell? 😉 …which I also fol­low con­ti­nously) and the MMT

    Another thing, back in Swe­den I believe we have a sim­i­lar sit­u­a­tion as you’ve got in Aus­tralia regard­ing an asset bub­ble (even if the Swedish ver­sion has been fueled by other means, mainly tax cuts on salaries and prop­er­ties). Have you ever had the oppor­tu­nity to study this and apply your mod­els?

  • Wel­come aboard Neb­bi­olo71,

    As for Swe­den, I’m sorry but that hasn’t turned up on my radar; once we get an online data­base rolling, hope­fully data will be avail­able for easy com­par­i­son across coun­tries, but I don’t have any for Swe­den yet.

    On Char­tal­ism ver­sus Cir­cuit The­ory, I have been hold­ing back from writ­ing a cri­tique for some time but will soon be embark­ing on one. I don’t dis­pute their account­ing of how gov­ern­ment financ­ing works, or the need for a gov­ern­ment deficit when the pri­vate sec­tor is fal­ter­ing. I do ques­tion their eco­nom­ics, which I see as being essen­tially sta­tic in nature (despite their argu­ments about stock-flow con­sis­tent mod­el­ing) and based on an erro­neous con­ser­va­tion law, their rel­a­tive non-con­sid­er­a­tion of the role of pri­vate sec­tor credit creation–which as you know is my main ana­lytic focus–and their ten­dency to treat the pri­vate sec­tor as an aggre­ga­tion, whereas I argue (from a Cir­cuit per­spec­tive) that at least three classes must be dis­tin­guished to con­sider pri­vate sec­tor dynam­ics prop­erly: work­ers, cap­i­tal­ists and bankers.

    Those are pre­lim­i­nary remarks only. I won’t pub­lish a detailed analy­sis and cri­tique until I’m ready.

  • Neb­bi­olo71

    Hi Steve and thanks for your prompt reply (from down-under?)!

    My sus­pi­cion of the ori­gin of the dis­pute, sta­tic ver­sus dynamic mod­el­ing, was some­what right then and I’m look­ing for­ward to read your com­ing pub­li­ca­tion of your more explicit cri­tique of the MMT model/Chartalism.

    From an Euro­pean per­spec­tive it’s extremely inter­est­ing and promis­ing to see that so much new think­ing is com­ing from down-under and I’m not only talk­ing about your (and Bill Mitchell’s) work but also work from peo­ple like Geoff Davies.

    Regard­ing Swe­den, it’s cur­rently con­sid­ered one of the Euro­pean tiger economies but per­son­ally I think the econ­omy is mostly fueled by the credit impulse injected by the still ongo­ing hous­ing bub­ble. Of course, in addi­tion, because of not being part of EURO, Swedish export indus­try has flour­ished when the cur­rency has been floating/adapting but this doesn’t explain all. 

    For years the increase in house­hold debt has been in the range of 6–8 % of GDP and is now reach­ing 100% in total (of GDP), total pri­vate debt clos­ing in on the 300% of GDP. The inter­est­ing is that this period of growth started by quite high unem­ploy­ment (cre­ated dur­ing the last Swedish finan­cial cri­sis in 1993 when the gov­ern­ment tried to tie the cur­rency to the EURO…) that hasn’t changed/improved despite the growth of GDP. The ratio between mean income and aver­age debt is sim­i­lar to the fig­ures you’ve pub­lished on Aus­tralia, maybe even a bit worse.

    What wor­ries me (an a whole bunch of oth­ers) is what hap­pens once the increase in house­hold debt declines or worse, if peo­ple start­ing to pay back loans instead of tak­ing on new ones. Read­ing your work on the sub­ject doesn’t really dis­pel these dark clouds… 😉 …when I think your analy­sis would point out a strong reces­sion or even depres­sion, given the mag­ni­tude of the debt (espe­cially if the world is on a pro­tec­tion­ism route, given the impor­tance of export in Swe­den) — am I right?

  • sir­ius

    Maybe Ben Bernanke will start read­ing your blog ?

    Bernanke Swan­Song ?

    a video with David Stock­man — I am only 2 min 50 into view­ing and it is scathing of Bernanke

  • Frank

    Here is an inter­est­ing one:

    First hand expe­ri­ence of false economies.

  • Those sound like cri­sis level fig­ures to me Nebbiolo–similar to the USA which is much worse than Aus­tralia on debt to GDP. And your export vul­ner­a­bil­ity sounds like a replay of Japan, which slumped very badly when the Great Reces­sion began since its exports plum­meted.

  • RickW

    This link gives the finan­cial data for Swe­den:
    I con­sider the net for­eign invest­ment posi­tion to be a good indi­ca­tor of the health of an econ­omy. I started look­ing at this a while back because it seems there is debt every­where but there must be credit in the sys­tem to bal­ance the debt.

    This data shows Sweden’s net posi­tion at Q1/2011 was –SEK628bn against a quar­terly GDP of SEK853bn or say –18% of annual GDP. Even bet­ter news is that it improved from the pre­vi­ous quar­ter.

    The US is worse than this but not all invest­ments are the same because US enti­ties get favoured treat­ment and earn higher income on loans than do for­eign­ers hav­ing assets in the US. How­ever US is still going back­wards at a blis­ter­ing pace.

    Aus­tralia is much worse at around –88% of GDP but was mak­ing good gains up till end of 2010 then hit by bad weather giv­ing poor BoP in Q1 2011. Some of the recent invest­ment in Aus­tralia has been in pro­duc­tive assets but a lot of that invest­ment comes from off­shore — notably China. In 2009 Aus­tralia was worse than Ire­land. Aus­tralians have a high pro­por­tion of their wealth tied up in the hous­ing Ponzi scheme that the Fed­eral Gov­ern­ment encour­ages in var­i­ous ways.

    I am baf­fled by the stu­pid­ity of econ­o­mists who fail to fac­tor in the sig­nif­i­cance of debt in an eco­nomic model. I have sig­nif­i­cant con­cern that the only way for­ward will be through infla­tion to devalue cur­rent debt and my sav­ings with it. The incom­pe­tents who encour­aged the debt get rewarded, those who enjoyed the ben­e­fits of the debt get for­given and those who worked long and hard to save are proven stu­pid. What a mess!

  • sir­ius

    News just come in…

    Greece???” says “yes”…to…austerity and more raids from __________?

    I always “love” intim­i­da­tion in our “demo­c­ra­tic sys­tems”…

    One of those who voted against it — a deputy of the rul­ing party — was imme­di­ately expelled by the Greek prime min­is­ter George Papan­dreou. ”

    What­ever the real (physics) energy sav­ings the pri­vate sec­tor makes you can count on “gov­ern­ment” to more than off­set them…

    The gov­ern­ment resorts to the old tried and tested favourite of “let’s build a road”.

    In this case a high speed rail link (dubbed “HS2”) for the “com­mon good” esti­mated to cost a mon­e­tary 33 bil­lion UK pounds.

    So what if we have to “destroy some GDP” in the form of houses, gar­dens and agri­cul­tural land that lay in the new path for this rail link?

    Ah but ’tis for the ‘com­mon good’” explains the bar­ris­ter who will now be able to travel to the North of the UK from Lon­don on the same day rather than begin his jour­ney the day before.

    (Do we really need bar­ris­ters from Lon­don?).

    As I pre­vi­ously stated dou­ble the work­ing speed of a mode of trans­port and you dou­ble the energy used (based upon the human view of what is the accept­able max­i­mum time for a jour­ney).

    All this in time of energy and resource con­straints.

    We have to dis­tin­guish between “good gov­er­nance” and “bad gov­er­nance”.

    The clown car accel­er­ates.

  • sir­ius

    Clar­ity for prior post.

    The high speed rail link is to be built in the UK.

  • alain­ton


    Hi, Im won­der­ing about the lead-lag issue — it does seem an unre­solved puz­zle.

    Im won­der­ing if a ‘table­top’ model might resolve it — the idea come from growth of small busi­nesses mod­el­ling.

    At any one time there are firms that are form­ing (com­ing onto the table­top) firms grow­ing, or declin­ing (slid­ing off) and dying (falling off). I wish I could remem­ber a ref­er­ence.

    Same with debt, new loans aris­ing, some grow­ing and com­pound­ing, some default­ing and some being paid off and paid off com­pletely.

    Invest­ment comes from accel­er­a­tion of credit for new loans.

    Exist­ing loans com­pound, if debt restricts firm bud­get con­straint too much then even accel­er­a­tion of credit wont go to invest­ment if go towards pay­ing off inter­est and inter­est on inter­est.

    So at times of very high debt a credit accel­er­a­tion can sim­ply go towards restor­ing bal­ance sheets, and when bal­ance sheets are restored firms are safe to accel­er­ate credit to expand — its a two way street — both lead and lag effects?

  • Brian Han­ley

    Saw on Kash Mansori’s blog that he says the USA’s banks have greater expo­sure to Greek debt than the euro­pean banks do. US Finan­cials have been sell­ing CDS’s on Greek debt to the Euro­peans. See:

    I have to ask the ques­tion, because it makes sense given the struc­ture of the “too-big-to-fail” game. Have both USA and EU financiers/bankers been con­spir­ing to stick the US tax­payer with the bill for the EU? One can model it as sim­ply a ratio­nal response to the game’s struc­ture. No player in the finan­cial sys­tem loses in that sce­nario, and it could explain the huge (and I think oth­er­wise utterly non­sen­si­cal) dif­fer­en­tial between EU and USA behav­ior rel­a­tive to Greek debt. 

    There is also the ques­tion as to exactly what the expo­sure really is, which is dis­cussed on the blog with­out res­o­lu­tion.

    How much of CDS lia­bil­ity is col­lat­er­al­ized? (i.e. Is there col­lat­eral, and to what degree does the col­lat­eral cover the posi­tion on paper?)
    And what, pre­cisely, is the form of the col­lat­eral? (i.e. To what degree is each col­lat­eral posted liq­uid? Is the col­lat­eral itself an inflated asset? And if it is traced back, is the same col­lat­eral pledged mul­ti­ple times over to cover dif­fer­ent CDS’s?)

    Unless you know the answer to those ques­tions, it is impos­si­ble to eval­u­ate, and as far as I know, there is sim­ply no source for such data.

  • Hi Andrew,

    I’ve got that feed­back from my more sta­tis­ti­cally capa­ble col­league, and I did get my leads/lags wrong–relying on my own rou­tines led me astray some­what. There are lead/lag rela­tion­ships which are unsta­ble but the rela­tion­ships are sta­tis­ti­cally sig­nif­i­cant. Once I’ve given the cur­rent post on Home­less­ness enough time for air, I’ll pub­lish my colleague’s results and the orig­i­nal data for oth­ers to exam­ine.

  • A. Kadl­cik

    Hello Steve,

    About neo­clas­si­cal econ­o­mists not read­ing their lit­er­a­ture. I wouldn’t be sur­prised about that. Now my expe­ri­ence with “higher learn­ing” is not on the research part of the equa­tion but on the study­ing part (human­i­ties, social sci­ences). I have been amazed time and again at the out­put of the so called research which has been empir­i­cally fal­si­fi­able, mean­ing­less (with­out mean­ing and/or incon­se­quen­tial), or inter­nally incon­sis­tent. (Not all has been bad, there have been also some true gems.)

    I have won­dered about that and my con­clu­sion is that there must be a silent “gentleman’s agree­ment” not to crit­i­cise each oth­ers’ works and thusly, keep the bulk of the acad­e­mia in busi­ness. (Most likely the not read­ing one’s lit­er­a­ture part is there as well because it is not at all unusual to read things claimed as new which, in the end, have been said before — sev­eral times — and more elo­quently.)

    Now, onto eco­nom­ics. Although here my “exper­tise” is very lim­ited, I have nonethe­less been under the impres­sion that quite a lot of eco­nom­ics — at least the kind which lay­men and the like are likely to hear about, at least from the main­stream sources — is based on one strange premise: the abil­ity to cre­ate a pre­dictable model of essen­tially a chaotic sys­tem (econ­omy).

    I mean, we have sophis­ti­cated math­e­mat­i­cal mod­els and for­mu­las — which by the way com­pletely sur­pass my com­pe­tency to under­stand them — yet, it seems, in order for them to work, they must assume the chaotic ele­ment, or ran­dom­ness, away. Such an act may indeed cre­ate a beau­ti­ful theory/model but if it is unfit to describe real­ity, why have it in the first place.

  • alain­ton

    Hi Steve,

    Just a hunch but ive been won­der­ing if the lead-lag issue might be closely related to a fun­da­men­tal cap­i­tal the­ory prob­lem.

    That is the prob­lem that you cant use the fisher stock-flow view of cap­i­tal and, at the same time, use mar­ginal pro­duc­tiv­ity the­ory to deter­mine the “opti­mal” level of cap­i­tal, and then have mar­ginal effi­ciency of invest­ment the­ory deter­mine the opti­mal level of invest­ment — you no longer have a flow invest­ment term, its overde­ter­mined. You can have one but not both the­o­ries — and Fisher seems to be right.

    Now to avoid this prob­lem neo-clas­si­cism devel­oped Tobins-q, as much loved by Lucas.

    But it doesn’t work — neo­clas­si­cal eco­nom­ics argues that the fluc­tu­a­tions in Tobin’s q are due to bub­bles and crashes which can be fore­casted by the mar­ket value of equi­ties and bonds mov­ing up faster that the cost of cor­po­rate fixed assets in a bub­ble, or drop­ping faster than the cost of cor­po­rate fixed assets in a crash.

    The trou­ble is that empir­i­cal work shows that exactly the oppo­site occurs (Nitzen and Bitch­ler 2010)

    This had already been pre­dicted by Hen­nesy (2004)
    because of debt over­hang. and Bolton, Chen and Wang (2009) who pre­dict an inverse rela­tion between mar­ginal Tobin’s q and invest­ment when the firm draws on its credit line.

    Just a hunch but i’m won­der­ing what hap­pens when you plug the accel­er­a­tion of debt into these debt over­hang equa­tions, in macro­eco­nomic terms. Can it explain the invest­ment cycle in a way which knocks the final nail in the cof­fin of the mar­ginal pro­duc­tiv­ity the­ory of cap­i­tal?