A Macroeconomics Debate at Cambridge
I gave the talk below to the Cambridge Society for Economic Pluralism yesterday. This student-formed society is attempting to open economics to debate–something which, despite the enormous schisms that exist within economics, is in practice sadly lacking.
Economists of one school of thought (such as the Neoclassical) don’t listen to or debate with those from others (such as the Post Keynesian or Austrian)-as you can see from Cochrane’s dismissive remarks about non-Neoclassical economics in the Playboy article on economics. Even within schools (such as the Neoclassical), different factions barely communicate with each other-as you can see by perusing some of the “Freshwater, New Classical” versus “Saltwater, Old Hicksian” (whoops, sorry, they think they’re “New Keynesians“) blog entries.
As well as inviting me to present on the Post Keynesian alternative macroeconomics that I’m developing, they invited Pontus Rendahl to provide a response from a more Neoclassical point of view (he described his approach as “Meterodox”, which was a rather clever phrase). Pontus’s response is linked here.
There were some points at which we were at cross-purposes–for instance, what he describes as my model was in fact Minsky’s model from 1963, and it therefore pre-dates by almost two decades the Lucas “Cash in Advance” model he later describes as pre-dating me (so rather than me being “nice but not novel” in relation to a 1982 paper from Lucas, Lucas was “nice but not novel” in relation to a 1963 paper by Minsky); I use differential equation notation for debt when working at the aggregate level and delta notation when considering a single transaction for technical reasons related to the aggregation of discrete asynchronous events, and so on.
But that’s OK: Pontus hadn’t encountered my work prior to accepting the student’s invitation to speak, and no-one can be expected to get completely on top of an alternative perspective at first try. What I appreciate is that he did engage, and I hope we’ll keep doing so occasionally from now on.
I’m sorry that I don’t have the time to edit the audio below–it starts with about 3 minutes of crowd noise (there were about 100 people in the audience). So please just skip about 3 minutes in where you’ll hear the introduction, and then my talk followed by Pontus’s, and finally the discussion.
There was quite a bit of heat directed Pontus’s way from some of the older academic staff in the audience who were involved in the internal battles that transformed Cambridge UK from a bastion of criticism of Neoclassical economics into a Neoclassical stronghold. Since Pontus only arrived at Cambridge in the last few years, this was all news to him, and he was justifiably rather taken aback.
Young members of the economic community can’t be expected to know this history, ironically because of one of the common complaints that I and other heterodox economists make about economics education today: it lacks any reference to economic history or the history of economic thought. But that’s the fault of the existing teachers, not of new entrants into the profession. It’s therefore little wonder that young Neoclassicals first encounter the hostility older non-Neoclassical economists feel, they can believe that heterodox economists are attempting a purge, when in fact they’re reacting to a Neoclassical purge that began in the 1970s and ended before most of them were born.
I’d rather forget those old conflicts and focus instead on the need to reform economics today, by acknowledging its empirical failures and by embracing the complex systems approach that has transcended equilibrium thinking in so many other disciplines.
Dear Steve,
Could I please ask a question. Beyond the next 5 years of deflating Real-estate prices, where do you see general prices of goods and services in the long term, say 10- 25 years forward?
Very high inflation, a collapse of the purchasing power of the global Fiat currency?
Steve, when will you start adding debt to income limits, 0 bound cash stock, bankruptcy for workers and businesses?
It seems to me that these 2 components: no more loans for individuals and companies deemed by banks to be high risk AND individuals and businesses must maintain greater than zero cash stock at all times, work in concert to precipitate an aggregate debt contraction.
If my thinking is correct, this is an important part of the system that turns the exponential debt growth into a crash. I am specifically refering to the plot @32:13 on the youtube video.
Finally, bankruptcy seems the method in which the whole system resets.
Of course, there are some people/businesses hitting the debt and cash limits all along, and reseting with bankruptcy.
Dear Steve
Quantitative Easing will surely debase the currency in the USA and ultimately will happen worldwide, will this cause massive inflation in the years to come. If yes, when? If not, how is that possible? Any help with my question is appreciated.
Steve
Would all the concerns about capital be irrelevant, since money can be “created out of thin air”?
http://www.zerohedge.com/news/2012-11-29/gold-solution-banking-crisis
Dear Professor Keen,
First of all, kudos to you for hammering away at the flaws in mainstream economics.
Your graphs and correlations clearly show that debt and the growth of debt are very important, but I have some nagging questions about your Walras-Schumpeter-Minsky law. I think you are right that new debt allows the economy to expand, but this shows up in a change in GDP from one period to the next, not in the concurrent measures apparently used in the WSM law.
As far as I can understand the documentation of the National Income and Product Accounts, cash flows from saving, borrowing and asset sales are already included, so it is redundant to add them in again if you are using the published standard values of Y and GDP. By careful construction, the value of Y is made equal to the concurrent value of GDP, so I do not understand how extra terms can be added to the simple equation: Y = GDP.
Here is how I look at it in a couple of simple cases:
Start off with an economy with a steady level of GDP = GDP0 = Y0 during a particular period of time. One agent, A, takes out a new bank loan of L dollars. If A does not spend the loan, the economy is unchanged (ignoring bank charges). If A does spend the loan on products from agent B, then B earns more and the total income goes up to Y0 + L, B’s production also goes up by L, so that GDP = GDP0 + L. So, when a bank loan is spent into the economy in a particular period, it already shows up in the measures of Y and GDP.
The story is a little different for loans from non-banks. If agent A borrows L dollars from another agent, C, and spends them, the GDP does not change, to first order, because A’s increase in spending offsets C’s decrease. There is no change at all if A spends the money the same way as C would have done, but there are second-order changes if A spends it differently and thereby alters the structure of the economy.
So it looks to me as if new money from bank loans does expand the economy, but in a way that already shows up in GDP. Loans from non-banks do not expand the economy, but they do have the potential to redistribute the activity.
If this is the right way to look at it, then rather than adding change of debt to Y, it might be better to follow Minsky and try to find a connection between the change in bank loans and the change in GDP over time.
QE is not QE Kalman: it is “PE”–Price Easing. The program involves Repo agreements where the Fed sells the purchased bonds back to the seller bank at a later date.
As I explain in an earlier post, QE reserves can’t be lent into the economy–only borrowed reserves can. So it doesn’t affect the money supply until banks are willing to borrow to lend.
And currency collapse is a many faceted process: if the US $ falls then yes imported inputs cost more, but they themselves can be subject to currency falls which neutralize the effect.
I do not expect hyperinflation out of this process, and any inflation I expect will be moderate towards deflation. The only inflationary forces I foresee are from global warming and peak oil.
Yes Drew, bankruptcy is such a mechanism–I likened it to Hawkins Radiation in the black hole analogy of a debt collapse. I just haven’t thought out a nice way to model that at a collective level yet.
I work in continuous time Blackbox, for a whole range of reasons. That eliminates the problems you foresee here which emanate from a discrete time, period approach.
Off the cuff–before reading zerohedge–no I don’t think they are irrelevant. Equity backing is needed for lending (though not reserves).
Mathematics, being a pure mental science is really just the opposite end of the monist actual reality that includes on the other end, Wisdom, which is the pure intuitive “science” that complements it (mathematics). Math is reductionistic, Wisdom wholistic. Both are absolutely essential and legitimate aspects of the actual one reality and yet there is a hierarchy of mental/psychological/spiritual (not supernaturally religious) reality. Why? Because the nature of Wisdom being in fact wholistic, continually binds it back to Math (science) and so makes it consciously aware of both realities. Math/science being reductionistic in its perspective has no such necessary thrust or inclination. It IS probably true that the best mathematicians have in fact drawn upon the wholistic, intuitive nature of Wisdom in their solutions. And that, I assert, is exactly what is most needed in economic, financiual and monetary systemic thinking. When Robert Reich says in the video on his Kickstarter sight for his documentary film Inequality For All, “The economy is for us, not us for the economy.” He precisely mirrors C. H. Douglas’s statement, “Systems were made for Men, not Men for systems, and the interest of man which is self-development, is above all systems, whether theological, political or economic.” Wisdom, the wholistic crown of Man’s achievement and potential, and the very accurate distinction made in his species designation, homo sapiens, is the inward experiencial guide whose universal, secular and empowering condensations of Confidence, Hope, Love and Grace are also the basis for the best temporal policy. As inside, so outside, as above, so below. Keep the “Faith”, baby!
Steve,
If you say “Equity backing is needed for lending (though not reserves)”, then how could money (or lending in your definition) be “created out of thin air”, in the sense of being unconstrained? How could excessive private debt (you harp on about) be created without sufficient equity backing?
By varying leverage ratios Lyonwiss.
Leverage ratios prevent abitrary credit “creation out of thin air”, which you have asserted (without qualification) many times in the past.
“How could excessive private debt (you harp on about) be created without sufficient equity backing?”
How about through the currently enforced fact of the rate of flow of total individual income scarcity in comparison to total prices from the moment of a product’s presence in the economy plus the all too natural tendency in a profit making system for its price to inflate right on through to retail sale? …..all of this endemic price inflation of course necessitating borrowing at an increasingly accelerating rate. Finance is more than happy to provide the additionally inflationary fact of this, their product. Additional money injection by central banks, government or exporting and any possible, if not probable and minuscule velocity of money are not solutions to this, just can kicking measures within the entire ponzi nature of the system from beginning to end.
The increasing debt burden on the system and the progressively austere loss of value of individual effective demand caused by the endemic inflation, and combined with the growth of a huge consumer market made possible by the increased productivity of technology which finance is again all too happy to exploit…..and individual debt follows.
So the cause of accelerating debt is both the nature of the present system itself….and the failure to correct the scarcity enforcing accounting flaw with a new consumer financial paradigm of a perpetual Dividend to bridge the gap between individual incomes and prices and a general discount to negate the total of the system’s natural inflationary tendencies.
The effect of these measures would be a huge actual decrease in the money needed to keep the economy flowing….because the additionally inflating costs of consumer borrowing would be much less necessary and tempting.
Not co-incidentally these policies reflect Grace. Wisdom, the way out, the way home.
Professor Keen,
Please keep sharing your efforts to educate and persuade members of the economics profession. As an engineer, my sense of the economics classes I took was that the models were much to simplified to be useful. Your work is proving that suspicion to be true.
Thank You for sharing,
MHR
Dear Steve,
Thank you for your reply, I really respect your opinion and I see your point.
May I ask how you think is the USA and other indebted countries are going to deal with the overwhelming debt they are carrying without inflation or default? Surely its easier to repay debt with worthless pieces of paper?
Also at some stage the banks will borrow, at a very low rate to lend out, this is the business there in, they have nothing to lose “too big to fail” and interest rates are likely to be super low so it must entice people to borrow also without any governance.
In any case if deflation is really set in for years, the government will just send checks to people for many thousands of dollars just to kick start the economy. They have done this before, surely they will do the same with large amounts in the future, perhaps a $100,000 per person or more, it cost the government nothing to make the FIAT money and it will buy votes. More people will be happy at the beginning and less people will lose the value of their savings since they have none.
Can I conclude from your theory that Gold and Silver are a misplaced investment now, since high inflation is not likely to break out in the near-term?
Kind regards, Kalman Radvanyi
I don’t see how continuous time avoids the double-counting issue, since the values of Y, etc, in any finite interval just come from integrating the point values over that interval. Unless your definition of Y or GDP is different from the conventional one, the components must add up to give the same answer, no matter how you you divide it up over time.
In any case, I believe that the published values of Y that you use in your plots must include the cash flows from saving and lending.
I do think that you are on to something important, but that it is about the change of Y (or GDP) from one time to another, a la Minsky, rather than the difference between simultaneous values of Y and GDP.
Blackbox, a point to remember is that not all spending results in an increase in assets. So GDP being purely a measure of monetary flow in a period won’t necessarily capture the whole economic picture. For example.
If I have $10,000 and I spend it on a new car from a manufacturer, then that $10K results in the creation of a new car. So the value of assets in the economy at the end of the process is larger by one car plus the money is available for spending or loan by the manufacturer or its employees and shareholders.
If I have $10K and use it to buy a plot of land there is no direct change in the value of assets in the economy but the money is available for spending or loan by the former owner or its employees and shareholders and thus still leads to an increase in the value of assets in the economy.
If I have $10K and spend it on buying your $10K debt from the bank then there is no direct change in the value of assets in the economy but the money is available for spending or loan by the bank or its employees and shareholders and thus still leads to an increase in the value of assets in the economy.
If I have $10K and use it to buy my own debt from the bank then the value of assets in the economy drops because my $10K loan, a financial asset, no longer exists. This will be partially offset by the bank making new loans and its employees and shareholders spending some of the $10K in the same way as in the first three cases.
So if we take the onward spending/loaning by the manufacturer/landowner/bank and any employees and shareholders out of the picture (since it is common to all four cases) we are left with case 1 in which the spending leads to an increase in the value of assets in the economy by $10K and no change in the amount of debt, case 2 and 3 which lead to no change in either and case 4 which leads to a reduction in the value of assets and debts by $10K
So the economic effects of spending will differ markedly even though in all four cases the contribution of my spending to GDP for the period is apparently the same.
Derek,
I agree completely that there is much more to the economy than the amount of spending. My argument was limited to the point that, as far as I can see, cash flows from borrowing/lending and asset sales are already included in the published values of GDP (or Y), so it does not seem appropriate to add them in again.
On the other hand, as Minsky pointed out, if you want GDP to increase from one time to another, you pretty much have to have an increase in debt, so that may be a fruitful direction to explore.
Cheers, BB. I think that’s a reasonable point of view. Prof Keen has demonstrated the empirical correlation and it may well be fruitful to look for the causal relation between changes in debt and the resulting changes in GDP. That’s why I thought it worth enumerating some cases.
Ahhhh…. I opened that Playboy link at work! Lucky no one was behind me at the time
V useful Steve that you highlighted Minksy point that increase in AD can occur either through an increase in Velocity (though I now prefer to talk about increases in turnover as it highlights economic actors rather than implying money having a motive force of itself)or newly created money, a point almost always missed.
A small clarification might be needed. It might be taken from your definition of savings as ‘negative borrowing’ that this is a linear function where excess of consumption over spending creates a demand for debt. This is a common mistake as even a judicious Janet might have a positive savings rate and still take out a large amount of debt to purchase ea large asset such as a house. Indeed Judicious Janets have good credit ratings. So consumption is a non linear function of income – savings + change in debt – debt repayment.
Ive commented on Pontus’s reply on my blog http://andrewlainton.wordpress.com/2012/12/06/some-notes-on-pontus-rendahls-review-of-keensian-economics/
As one “of the older academic staff in the audience who were involved in the internal battles that transformed Cambridge UK from a bastion of criticism of Neoclassical economics into a Neoclassical stronghold”, I appreciated your vigorous attack on neoclassical dogma. I had given up even having a debate with my colleagues in Cambridge.
I sympathise with your approach in criticising the dogma, although I would be even more openly critical, e.g. of Pontus, in accusing them of clinging to theories just because of their investment in the calculus and their wish to climb the professional economics ladder. I have developed (since 1976 with colleagues in Cambridge) dynamic simulation multisectoral models of the UK, European and global economies, designed to track the observed performance of the economies over time (for better or worse!).
I do take issue with the resolutely aggregate economics of both the Keynesians and the Post-Keynesians, since important economic effects take place at the disaggregate (e.g. industrial) levels that affect the aggregates. And I also embrace applied econometrics (which was rejected by the Keynesians and most Post Keynesian and by Keynes in his famous interchange with Tinbergen) as a means of bringing data to bear on theory, otherwise we could make up anything we like and cloth it with statistics, as the CGE modellers do when they use one year’s data to project 100 years into the future.
Intrinsic and central to these ideas is a theory of money which is more comprehensive and better grounded than earlier theories and I believe it provides a better explanation for the post 2007 financial crisis and what to do about it. However I am not convinced that complexity theory can help us very much, apart from emphasising our uncertainty about explanations and outcomes.
I am very much interested in MMT, Post Keynesian and Full Reserve Banking. I have been a long time reader of Keen, and before that, of Fisher. I am trying to help understand how these views relate and help educate the general public here http://open.salon.com/blog/clintballinger/2012/12/17/post_keynesianism_mmt_100_reserves_project_question_1
(I would love or any pro MMT and / or pro Chicago Plan /Full Reserve Banking people to help out in posts or comments.)
Prof. Keen, I have been trying to promote the Minsky project and its kickstarter where I can. And I am glad you were speaking at Cambridge, I was there for doctoral studies for a number of years, miss it at times.
Cheers,
Clint Ballinger