economic situation is dire

[blandy]

The only thing fiscally, seems to be to encourage business to spend their surplus, but how do you do that?

Tax breaks on capex spending is probably about the only thing you could do, but the electorate isn't going to like that when everyone else seems to be getting hit harder.

The problem is companies never want to spend money, they only spend money when they absolutely have to....

Is that really true? - I suppose it depends on what you mean by "have to". Many Co.s spend on R&D, on training, on new machinery and premises etc. but many do so from borrowing, rather than by raising money from shareholders or using capital.

They do it for future profits, I suppose, but whether they "have to" is a matter of opinion. In a recession they stop, in a upswing they seem to o it a lot more, which tends to indicate it's not really a need, more a "good idea" as long as they can forsee a return coming from the spend.

But as has been reported for a good while, despite low interest rates, Banks won't lend, or are making it very hard to borrow. They were required to lend when they were bailed out, but are barely if at all meeting their minimum obligations.

Gov't seems to have required them to do two contradictory things at once - lend more and increase their balance sheets.

Definite pickle, as you say

[peterms]

The only thing fiscally, seems to be to encourage business to spend their surplus, but how do you do that?

Tax breaks on capex spending is probably about the only thing you could do, but the electorate isn't going to like that when everyone else seems to be getting hit harder.

The problem is companies never want to spend money, they only spend money when they absolutely have to....

Is that really true? - I suppose it depends on what you mean by "have to". Many Co.s spend on R&D, on training, on new machinery and premises etc. but many do so from borrowing, rather than by raising money from shareholders or using capital.

They do it for future profits, I suppose, but whether they "have to" is a matter of opinion. In a recession they stop, in a upswing they seem to o it a lot more, which tends to indicate it's not really a need, more a "good idea" as long as they can forsee a return coming from the spend.

But as has been reported for a good while, despite low interest rates, Banks won't lend, or are making it very hard to borrow. They were required to lend when they were bailed out, but are barely if at all meeting their minimum obligations.

Gov't seems to have required them to do two contradictory things at once - lend more and increase their balance sheets.

Definite pickle, as you say

Companies will invest when they believe it is profitable to do so. At a time of reducing demand, with households trying to reduce their indebtedness, with governments cutting spending and talking up the need to impose austerity (ie reduce demand still further, making people fearful of losing their jobs and making them even more cautious), there is no immediate prospect of aggregate demand rising.

It's not so much about borrowing. Big businesses can still borrow, but smaller ones find it harder. The real problem is demand. That's why a policy approach which just focuses on keeping interest rates low and ignores the need to create more demand, isn't working. The notion that keeping interest rates low will stimulate borrowing which will create investment and lead us out of recession is something plucked from the pages of neoclassical textbooks. As Koo has explained (linked previously), in a balance sheet recession where people and companies are trying to reduce indebtedness, that won't work.

The answer is that governments need to spend. Doing that would stimulate the economy, create additional demand, and create the conditions where companies see opportunities for profitable investment. That governments won't do it is both ideological blindness, and a dereliction of their duty.

[peterms]

A slightly different view in the article below from the circular spending vs austerity debate, basically saying the problem is due to as yet undeclared losses in the banking sector, and their selfish efforts to maintain fantasy balance sheets that are in turn preventing a recovery.

UK and Europe languish in a 'zombie bank’ malaise

If that's correct then it all comes back to the wrong moves being made in 2008 when the banks were bailed out instead of being allowed to go to the wall.

It's correct, and people like David Malone have been explaining this for some years now.

But there is a very important connection with austerity. It is that the bailout is not a one-off event in 2008, it's a continuing state of affairs. The IMF loans to Greece are being paid directly to German banks, to protect them from the consequences of their own bad business decisions, and the Greek people are being asked to pay the cost of this through austerity rather than the banks write down the bad debts like any other business in the world would have to.

Same across Europe. The process of socialising bad bank debts, based on fraudulently overvalued assets (and covered up by tame auditors) has been running for several years, and shows no sign of slowing down. Yet every few months, we find more previously unannounced multi-billion-pound scandals, which we are told for some reason we will have to pay for.

[DeepDish]

Someone mentioned Steve Keen.

He is also mentioned by Chris Martenson in this blog.

He talks about the correlation between debt and house prices. I am 100% certain that a lot of the wealth development in the Western world the last years has been debt driven, and that we will go back to some level between now and the levels before the debt increses began.

I found one of Keen's suggestions interesting:

Give central bank printed money to the people instead of directly to the banks (where they are used to speculate in stocks, buying government bonds and used to fuel oil and fuel prices). The people have to use the money to pay down on debt, though.

An interesting suggestion, that is far better than handing it to big banks that just grow bigger. If the money are used to pay down debt, the banks will become smaller, and pose less risk for the financial system. Not sure how good it is, though.

I think the mistake [central banks] are going to make is to continue honoring debts that should never have been created in the first place. We really know that that the subprime lending was totally irresponsible lending. When it comes to saying "who is responsible for bad debt?" you have to really blame the lender rather than the borrower, because lenders have far greater resources to work out whether or not the borrower can actually afford the debt they are putting out there.

They were creating debt just because it was a way of getting fees, short-term profit, and they then sold the debt onto unsuspecting members of the public as well and securitized their way out of trouble. They ended up giving the hot potato to the public. So, you should not be honoring that debt, you should be abolishing it. But of course they have actually packaged a lot of that debt and sold it to the public as well, you cannot just abolish it, because you then would penalize people who actually thought they were being responsible in saving and buying assets.

Therefore, I am talking in favor of what I call a modern debt jubilee or quantitative easing for the public, where the central banks would create 'central bank money' (we cannot destroy or abolish the debt, which would also destroy the incomes of the people who own the bonds the banks have sold). We have to create the state money and give it to the public, but on condition that if you have any debt you have to pay your debt down -- no choice. Therefore, if you have debt, you can reduce the debt level, but if you do not have debt, you get a cash injection.

Of course, this would then feed into the financial sector would have to reduce the value of the debts that it currently owns, which means income from debt instruments would also fall. So, people who had bought bonds for their retirement and so on would find that their income would go down, but on the other hand, they would be compensated by a cash injection.

The one part of the system that would be reduced in size is the financial sector itself. That is the part we have to reduce and we have to make smaller. That is the one that I am putting forward and I think there is a very little chance of implementing it in America for the next few years not all my home country [Australia] because we still think we are doing brilliantly and all that. But, I think at some stage in Europe, and possibly in a very short time frame, that idea might be considered.

[TheDon]

http://www.youtube.com/watch?feature=player_profilepage&v=s-Qg-2d50Eo.

[Enda]

Please stop listening to Steve Keen. He does not know what he's talking about.

[TheDon]

Please stop listening to Steve Keen. He does not know what he's talking about.

Do you want to back that up with a counter argument showing what he's got wrong?

[peterms]

Please stop listening to Steve Keen. He does not know what he's talking about.

So where you say

Most economics courses involve a downward line intersecting with an upward sloping line, and calling the intersection "the equilibrium". I teach this stuff to very smart university students and they struggle with that.

do you mean IS-LM?

Where Keen discusses why IS-LM doesn't work, quoting the inventor of the model himself (Hicks) acknowledging that, what do you make of that? It seems to me he knows very well what he's talking about, has researched it extensively, and has explained his view and the reasoning behind it extremely fully.

Is it not a bit strange that university courses continue to teach a model of which its own inventor said

‘I accordingly conclude that the only way in which IS-LM analysis usefully survives—as anything more than a classroom gadget, to be superseded, later on, by something better—is in application to a particular kind of causal analysis, where the use of equilibrium methods, even a drastic use of equilibrium methods, is not inappropriate…

(Hicks, referenced in the Keen article).

Perhaps the reason the students struggle with it is that it is so detached from what actually happens in the real world. Which is Keen's wider point about how money is created, what effect that has on debt, and how that affects the economy; which is, he says, also misleadingly described in (most) university courses. And which in turn is why the advocates of models based on such flawed reasoning didn't suggest that a crash would happen, where those who weren't so bound up in simply repeating what they had been taught were able to make what has been shown to be a far more accurate assessment of what was happening.

It's a funny old world, where someone can accurately predict, with fully worked out reasoning and explanation, a massively important event and be decried as not knowing what they are talking about, by those who didn't see it coming and whose models still can't explain why it happened.

[Enda]

Please stop listening to Steve Keen. He does not know what he's talking about.

Do you want to back that up with a counter argument showing what he's got wrong?

Of course. It's well-trodden ground, so I'm going to reference other people's critiques too.

One of the dominant frameworks in macroeconomics is Dynamic Stochastic General Equilibrium (DSGE) models. Personally I don't like them very much at all, but I at least know what they are and understand what's going on. As Paul Krugman notes, Steve Keen doesn't understand what's going on in DSGE models. Under the most generous interpretation of this debate, Keen doesn't seem to understand so-called New Keynesian modelling.

Have a gander at Keen's wiki page. It claims Keen's work "has also focused on refuting the neoclassical theory of the firm, which argues that firms will set marginal revenue equal to marginal cost. Keen notes that empirical research finds real firms set price well above marginal cost: they charge a markup, often cost-plus pricing." There are many things wrong with this. First of all, setting marginal revenue equal to marginal cost is *not* the same thing as setting price equal to marginal cost. In only crazy theoretical examples (i.e. perfect competition) does price equal marginal revenue. In other cases, it does not. What's more, in the New Keynesian models that Keen was complaining about in the previous paragraph, it is often assumed that firms charge 'cost-plus' pricing. This sort of stuff is pretty elementary - it's taught to undergrads. For example, see equation 7 on page 4 of these Trinity College undergrad notes here.

As it happens, the author of those notes has commented on Steve Keen's work. The summary is that his work is an odd, faux-scientific restatement of comments made others with better understanding than Keen.

In the same vein, John Quiggin calls him to task here. For example, "Keen generally uses the term 'economics' to refer to the simplified version of neoclassical economics taught to first year undergraduates." So either he's unaware of the grown-up stuff, or he avoids dealing with it. Even while restricting it to the simple version of economics that is distilled to 18 year olds, he screws it up. As Quiggin says, "[he] is simply wrong."

All of this leads to his "most celebrated" work, downloadable here. Here he makes the remarkable claim that the "optimal level of strategic interaction between competing firms is zero", i.e. they make higher profits if they do not compete. No shit, Sherlock. The problem, of course, is that in his model, each firm has the incentive to cheat on that agreement. Strange how he fails to mention this, but this is obvious to anyone with a decent training in economics. This is why it's published in a physics journal and not an economics one. He's a bullshit peddler, pure and simple, and it's a pity he has risen to prominence in intelligent non-specialist circles.

So where you say

Most economics courses involve a downward line intersecting with an upward sloping line, and calling the intersection "the equilibrium". I teach this stuff to very smart university students and they struggle with that.

do you mean IS-LM?

No.

Where Keen discusses why IS-LM doesn't work, quoting the inventor of the model himself (Hicks) acknowledging that, what do you make of that?

I have a few thoughts on it.

1. The IS-LM model is almost a century old and is based on monetary rules that stopped existing decades ago.

2. Accordingly IS-LM does not enter any economist's head when they think about the macroeconomy. Modern macro was invented in 1936. IS-LM was first published (iirc) in 1937. It's pre-history, and has been completely superseded by AS-AD models, Solow-type models, DSGEs, etc etc.

3. Keen isn't breaking any ground when criticising the teaching of IS-LM. He's simply joining the debate; see here, here, here, or here. I'm sure Keen knows IS-LM; that doesn't excuse his other howlers.

4. I don't think it should be taught to undergrads. I don't think it's taught in the university where I work.

5. I don't object to teaching IS-LM to undergrads as a stepping stone to understanding something more reasonable. This is the logic of the author of the leading undergrad textbook.

[peterms]

One of the dominant frameworks in macroeconomics is Dynamic Stochastic General Equilibrium (DSGE) models. Personally I don't like them very much at all, but I at least know what they are and understand what's going on. As Paul Krugman notes, Steve Keen doesn't understand what's going on in DSGE models. Under the most generous interpretation of this debate, Keen doesn't seem to understand so-called New Keynesian modelling.

Did you read the comments below that short piece? It's striking how many of them criticise Krugman for a lazy, inaccurate critique of Keen based on leaving out (or probably not having read in the first place) the immediately following bit where Keen specifically talks about the sticky prices which Krugman says he doesn't acknowledge. Several of them suggest that a retraction and an apology by Krugman would be in order. They mostly appear to be people generally sympathetic to Krugman, who think he's simply wrong in what he says here, and patronising and high-handed in his lofty dismissal.

If you look at "Debunking Economics", Keen deals with DSGE models and explains why he finds them wanting.

Have a gander at Keen's wiki page. It claims Keen's work "has also focused on refuting the neoclassical theory of the firm, which argues that firms will set marginal revenue equal to marginal cost. Keen notes that empirical research finds real firms set price well above marginal cost: they charge a markup, often cost-plus pricing."

Seriously? You're criticising Keen based on a sentence someone has written about him on Wiki?

I can see why Krugman dislikes Keen so much - they are both bitchy about each other, and Keen doesn't endear himself by speaking of Krugman as a dog walking on its hind legs, stating that he has written about Minsky without understanding him, saying that were he Keen's student he would have failed him for inadequate work, and so on. But from what I've seen of the debate, Keen appears to have done far more by way of setting out what Krugman has said and then pulling it apart, where Krugman comes across as not engaging with the arguments, and trying to simply dismiss him like the lord of the manor brushing away the tradesman. It does him no favours.

[Enda]

One of the dominant frameworks in macroeconomics is Dynamic Stochastic General Equilibrium (DSGE) models. Personally I don't like them very much at all, but I at least know what they are and understand what's going on. As Paul Krugman notes, Steve Keen doesn't understand what's going on in DSGE models. Under the most generous interpretation of this debate, Keen doesn't seem to understand so-called New Keynesian modelling.

Did you read the comments below that short piece? It's striking how many of them criticise Krugman for a lazy, inaccurate critique of Keen based on leaving out (or probably not having read in the first place) the immediately following bit where Keen specifically talks about the sticky prices which Krugman says he doesn't acknowledge. Several of them suggest that a retraction and an apology by Krugman would be in order. They mostly appear to be people generally sympathetic to Krugman, who think he's simply wrong in what he says here, and patronising and high-handed in his lofty dismissal.

If you look at "Debunking Economics", Keen deals with DSGE models and explains why he finds them wanting.

Have a gander at Keen's wiki page. It claims Keen's work "has also focused on refuting the neoclassical theory of the firm, which argues that firms will set marginal revenue equal to marginal cost. Keen notes that empirical research finds real firms set price well above marginal cost: they charge a markup, often cost-plus pricing."

Seriously? You're criticising Keen based on a sentence someone has written about him on Wiki?

I can see why Krugman dislikes Keen so much - they are both bitchy about each other, and Keen doesn't endear himself by speaking of Krugman as a dog walking on its hind legs, stating that he has written about Minsky without understanding him, saying that were he Keen's student he would have failed him for inadequate work, and so on. But from what I've seen of the debate, Keen appears to have done far more by way of setting out what Krugman has said and then pulling it apart, where Krugman comes across as not engaging with the arguments, and trying to simply dismiss him like the lord of the manor brushing away the tradesman. It does him no favours.

Overlook the fact that it's Krugman and look at the substance of the points. Keen says that barter systems and perfect competition are "underlying principles" of DSGE models. That's just not true. I have seen DSGE models with financial sectors/money markets, for starters. I have seen DSGE models that don't have well-adjusting equilibrium properties, i.e. where government interference improves things. I have seen DSGE models with market power. And the entire basis of New Keynesian economics is the inability of firms to set prices "well" and where central banks can then play a role, i.e. you're not dealing with perfectly adjusting perfectly competitive markets acting seamlessly in a barter economy; you're dealing with the opposite. Thus I conclude that either he doesn't get DSGE very well or he really doesn't get New Keynesianism. This is compounded by the fact NK models aren't even necessarily DSGE models. (I admit that Krugman is very unclear on this point.)

I know why DSGE models are wanting. I'd argue, even based only on the above, that I understand their failures at least as well as Keen does. If you really want to critique DSGE look no further than Cogley and Nason's article in the American Economic Review, published in roughly 1995. That's a serious, well-founded, specific critique of DSGE modelling. Keen is hand-waving, and mixing things up.

I used the wiki sentence as an example of Keen screwing things up. Maybe the wiki sentence is wrongly attributed to Keen and in that case then I of course withdraw the comment. But it's an example among others such as the article I linked to in a physics journal where he does something that might appear as useful to the untrained eye, but ultimately is misleading.

[peterms]

Overlook the fact that it's Krugman and look at the substance of the points. Keen says that barter systems and perfect competition are "underlying principles" of DSGE models. That's just not true. I have seen DSGE models with financial sectors/money markets, for starters. I have seen DSGE models that don't have well-adjusting equilibrium properties, i.e. where government interference improves things. I have seen DSGE models with market power. And the entire basis of New Keynesian economics is the inability of firms to set prices "well" and where central banks can then play a role, i.e. you're not dealing with perfectly adjusting perfectly competitive markets acting seamlessly in a barter economy; you're dealing with the opposite. Thus I conclude that either he doesn't get DSGE very well or he really doesn't get New Keynesianism. This is compounded by the fact NK models aren't even necessarily DSGE models. (I admit that Krugman is very unclear on this point.)

Well, he distinguishes two broad schools of DSGE, and characterises the NK version as recognising imperfect competition. But he says that in his view, where adherents of NK see their ideas as quite different from the other school, he sees the variation as relatively trivial (which is why he calls them neo-classical, where they don't at all see themselves as such). I'm not sure if this conflicts with what you are saying, or if you agree this is what he says and disagree with his assessment.

But the larger point is surely this, and I keep coming back to it. The NK models, like the other DSGE models, like mainstream neo-classical thought, like the much-read Mr Mankiw, did not predict the crash, while other writers who paid more attention to the part played by debt in the economy did.

That much, it seems, is not in dispute.

And starting from that observation, to hear the ones who didn't see it coming telling the ones who did that they know nothing and that no-one should pay them any attention, leaves me and many other people feeling that they have to all intents and purposes lost contact with reality. It would surely be more productive as well as more honest to consider why their models didn't work, and what they should do about it.

[Enda]

I would agree that NK is essentially neo-classical with frictions. That's an accurate assessment, and I think most New Keynesians would accept that too. I disagree that the tweaks are trivial. (Though I would be generally sympathetic to the view that small frictions like status quo bias have a big effect on the world.) The most pertinent example of this non-triviality is that there is a role for central banks to affect money supply in a simple NK model, but not in a simple neo-classical model. That's very significant when you want to consider the effects of money/debt/finance.

You're right that he was correct on the role of debt, and in pointing towards Minsky in that regard. That's his contribution, and well done to him. But he has demonstrated an enormous lack of mastery of what everyone else is doing. He regularly makes howlers (e.g. perfect competition and no money as "principles" of DSGE; general equilibrium theory being "false" unless preferences are of Gorman form) so you need to discount his credentials.

What's more, prediction itself should be treated a little bit cautiously in economics. I'm not saying this to defend economists' ability to predict. Not at all. The second reason is the endogeneity/self-fulfilling nature of predictions. As I said earlier if we could confidently predict that stock prices would rise by 25% in 2013, the price shouldn't wait around and should rise tomorrow. Nobody knows when this consensus might form to inspire changes, so economic predictions are always going to be influenced somewhat by luck. The second point, as Nassim Taleb very nicely points out in 'Fooled by Randomness', is that society vastly over-rates those that turn out lucky. What matters at least as much as a previous ability to predict is a consistency in your analysis that will encourage future correct predictions. Economics also has one season wonders. The fact that Keen is all over the shop means you should at least consider the potential that he is a Francis Jeffers.

[peterms]

I would agree that NK is essentially neo-classical with frictions. That's an accurate assessment, and I think most New Keynesians would accept that too. I disagree that the tweaks are trivial. (Though I would be generally sympathetic to the view that small frictions like status quo bias have a big effect on the world.) The most pertinent example of this non-triviality is that there is a role for central banks to affect money supply in a simple NK model, but not in a simple neo-classical model. That's very significant when you want to consider the effects of money/debt/finance.

So do DSGE models see money supply as affected only by central banks, or do they see the banking sector as a whole having a crucial role in this? My impression is that they don't look at money creation by banks as a whole, where the "heterodox" economists are more inclined to view money as being endogenously created by banks, with the central bank in effect forced some weeks later to provide the increased reserves which have been made necessary by the banks' creation of new credit. And that scale of new money creation (created as debt) far outstrips anything the central bank typically does.

As I said earlier if we could confidently predict that stock prices would rise by 25% in 2013, the price shouldn't wait around and should rise tomorrow. Nobody knows when this consensus might form to inspire changes, so economic predictions are always going to be influenced somewhat by luck.

Yes they are, and even economists who correctly foresee a crash are unlikely to be able to say just when it will happen. If they could, no doubt they would be making lots of money. But some predicted a big crash caused by excessive and unmanageable debt, and many just didn't at all. The fact that people can't predict the timing of the crash is to me less significant than the difference between predicting it or not.

The second point, as Nassim Taleb very nicely points out in 'Fooled by Randomness', is that society vastly over-rates those that turn out lucky. What matters at least as much as a previous ability to predict is a consistency in your analysis that will encourage future correct predictions. Economics also has one season wonders. The fact that Keen is all over the shop means you should at least consider the potential that he is a Francis Jeffers.

Yes, society does very much overrate those who turn out lucky, and this is especially the case in the finance sector. But what I see is some economists who utterly failed to predict a crash writing off those who did with remarks like "He's predicted eleven of the last two slumps" and the like. There's quite a difference between running around shouting "We're all doomed" and explaining why the dynamics of seemingly limitless credit creation must lead to a collapse in the system. And a still bigger difference between that and not predicting it at all.

What do you think of Fred Harrison, and his idea that there is a cycle to the property market caused by the interaction between compound interest at historically typical rates, and house prices, so that there is a (roughly) 18-year cycle of boom and bust in the housing market?

[peterms]

...and a further interesting discussion here covering the limitations of microfounded DSGE models; including their divergence from reality, which, were economics really a science, would by now have exposed them to more criticism.

And following on from the author's comment about what central banks do in dealing with a conflict between theory and reality, there's a more recent paper here about the way the BofE is reviewing its approach to risk and uncertainty, with references to the role of theory in explaining (or rather not adequately explaining) the actual risks being run. Since it's written by the director of financial stability, it seems likely it's the product of some reflection on the failure of the bank to predict and manage the degree of systemic risk facing us.

It's also quite possibly the only speech by an executive director of the BofE to reference Kim Kardashian. Is that a good thing? I don't know.

[Enda]

So do DSGE models see money supply as affected only by central banks, or do they see the banking sector as a whole having a crucial role in this?

DSGE can do whatever you want them to do. Say you want a model with endogenous capital, labour supply, and output. Then you need to specify three equations to link them together. If you want to add a housing market, you need to specify another equation linking it into the system. You're welcome to do whatever you want to include a banking sector. It's all very ad hoc, but at least you can add as many bells and whistles as you like.

What do you think of Fred Harrison, and his idea that there is a cycle to the property market caused by the interaction between compound interest at historically typical rates, and house prices, so that there is a (roughly) 18-year cycle of boom and bust in the housing market?

Never heard of him if I'm honest. I'd be suspicious that it's spurious, since these cycles should be unpredictable and every fibre in my body tells me that something so complex can't possibly be captured that simply. (Wouldn't it be great if it could, though?)

The other links you provide are great, thanks. I'd much rather if the likes of Simon Wren-Lewis got media attention at the expense of the likes of Steve Keen. Haven't seen any economics paper reference Kim Kardashian before, but I do know that one economist in Berkeley tries to sneak "Johnny Depp" into each of his papers.

[Risso]

As interesting as it is reading Economic Theory For Beginners every evening, if I really wanted to do that I'd go into the loft and dredge up my university notes.

[Jon]

GF Stanlake FTW!!

Takes me back to A Level days, sadly

[TrentVilla]

As interesting as it is reading Economic Theory For Beginners every evening, if I really wanted to do that I'd go into the loft and dredge up my university notes.

Well if you do get the ladder down you might as well stick those on e-bay, I hear Peter Pannu is on the look out for some.

[peterms]

So do DSGE models see money supply as affected only by central banks, or do they see the banking sector as a whole having a crucial role in this?

DSGE can do whatever you want them to do. Say you want a model with endogenous capital, labour supply, and output. Then you need to specify three equations to link them together. If you want to add a housing market, you need to specify another equation linking it into the system. You're welcome to do whatever you want to include a banking sector. It's all very ad hoc, but at least you can add as many bells and whistles as you like.

No, I don't mean can DSGE models include a banking sector, I mean isn't it the case that they tend to overlook the role of debt because they are based on the idea of exogenous money creation and see debt as someone else's asset rather than something which changes the size and nature of the economy, therefore making no great overall difference; while those who see money as endogenously created by banks see the rapid increase in the money supply (as debt) as a crucial factor which DSGE theories don't recognise (whether or not they include banking in their models). Though I have seen comments to the effect that Krugman's position on endogenous/exogenous money is unclear, eg here.

What do you think of Fred Harrison, and his idea that there is a cycle to the property market caused by the interaction between compound interest at historically typical rates, and house prices, so that there is a (roughly) 18-year cycle of boom and bust in the housing market?

Never heard of him if I'm honest. I'd be suspicious that it's spurious, since these cycles should be unpredictable and every fibre in my body tells me that something so complex can't possibly be captured that simply. (Wouldn't it be great if it could, though?)

Basically saying that there is roughly a 14 year cycle to house price growth, becoming more frenzied and speculative, with a couple of years of crash then readjustment, so broadly an 18 year business cycle (thrown out by eg world wars, but fairly constant apart from things of that magnitude). He claims that the evidence bears out the existence of such a cycle over some 400 years, but that there has been little attention to it because economists have focussed on labour and capital more than land (and it's land price speculation rather than building costs which is happening here). He says this particular market is linked to the cycles in the broader economy partly because of its sheer size and importance, and partly because the requirement to put up land as collateral for loans creates a link between escalating land values and the ready availability of credit, including for the very speculation which fuels the house price spiral.

Harrison predicted the crash in a book published in 2005, which is actually earlier than people like Godley, Roubini, Baker, Keen and others started talking about it.