Economyths - Book Recommendation

[tomandlu]

Just read "Economyths" by David Orrell. Highly recommended...

https://en.wikipedia.org/wiki/Economyths

Quote

The economist Robert Nelson wrote in the International Journal of Social Economics that the book suffered from a number of omissions, in particular a fuller exploration of “the religious roles played by neo-classical economics.” However he concludes that “Whatever its omissions and other failings, much of the book is devoted to making a strong case for one very important finding – the intellectual poverty of neo-classical economics … A wide audience including many non-economists could benefit from reading it.”

Lots of classical economists seem to hate it - I take that as a mark in its favour...

[zugzwang]

8 minutes ago, tomandlu said:

Just read "Economyths" by David Orrell. Highly recommended...

https://en.wikipedia.org/wiki/Economyths

Lots of classical economists seem to hate it - I take that as a mark in its favour...

I read the original in paperback 8/9 years ago, about the same time that I discovered Steve Keen. It's easier to digest than 'Debunking Economics' but every bit as solid. I'd also recommend 'Apollo's Arrow' and 'Truth or Beauty' by Orrell which are respectively concerned with the limits of certainty and how the quest for mathematical elegance in science is frequently misleading. Both issues of huge significance in the institutional and operational failure of economics as a research program.

[Will!]

I knew some Economics students at Warwick (which had/has a highly-regarded Economics department, if that means anything).  They described much of Economics as being like theology, in which charismatic individuals develop cult-like followings, heretics are ostracised and some debates are the equivalent of how many angels can dance on the head of a pin (“In Economics the real world is often a special case.”)

[scottbeard]

Sounds an interesting book, I might check it out.

Economics at the moment seems to be getting something of a bad name, which as an Economics graduate (of the 1990s) I find a bit sad.  Economics - at least as I was taught it - is not (and does not purport to be) a true science like maths or chemistry: there can't be any universal fundamental laws of Economics when you're dealing with people instead of molecules.  Yes in Economics 101 concepts are first taught and explored for cases where all people act rationally, or all have perfect information etc but of course everyone knows that this isn't the case in real life: that doesn't mean Economics is rubbish, just as Maths isn't rubbish just because at primary school children start by only doing basic simple things like "John is sharing out 9 sweets" instead of jumping straight to modeling complex real-life problems.

Economics provides the framework to help answer questions like:
- If you had £100 million to invest in education, would that provide more value spent in primary schools, secondary schools or universities?
- If we raise top rate income tax from 45% to 55% would that raise more money or less?
- If I want to increase my business's profit for next year, am I better to spend money on more marketing or more research and development?
- If we nationalized the UK energy market, who would be the winners and losers and by how much?

But Economics cannot actually tell you which is the "best" course of action in any of those scenarios, usually because any decision is not universally positive but rather has winners and losers, so only your beliefs can make that judgment call.  What Economics does is give some numbers that you can use when making Philosophical or Political decisions (hence why Oxford teaches the three together as PPE).

For example, I can't tell you whether it's "better" or "fairer" for the top rate of tax to be 55%.  But I could look at historical data on how tax take has changed when taxes have been changed in the past, and use that to help you understand what the financial and behavioral impact of the change might be.  Whether it's then the "right" thing to do is up to you.

Now of course, individual economists have individual viewpoints, and will articulate the course of action that they personally believe is best - but I think that's to be expected given there are no true "laws" of economics, and not a failing.

 

[Dorkins]

1 hour ago, scottbeard said:

Economics provides the framework to help answer questions like:
- If you had £100 million to invest in education, would that provide more value spent in primary schools, secondary schools or universities?
- If we raise top rate income tax from 45% to 55% would that raise more money or less?
- If I want to increase my business's profit for next year, am I better to spend money on more marketing or more research and development?
- If we nationalized the UK energy market, who would be the winners and losers and by how much?

But Economics cannot actually tell you which is the "best" course of action in any of those scenarios, usually because any decision is not universally positive but rather has winners and losers, so only your beliefs can make that judgment call.  What Economics does is give some numbers that you can use when making Philosophical or Political decisions (hence why Oxford teaches the three together as PPE).

Okay, but mainstream macroeconomics totally failed to see 2007-8 coming. That was not a normative question of 'what is the best policy', it was a modelling question along the lines of 'given these macroeconomic conditions and this set of government and central bank policies, what will happen next?'

There are schools of macroeconomic thought that claim to have predicted 2007-8 e.g. followers of Minsky like Steve Keen or the Austrians, but they are on the fringes. Mainstream macroeconomists don't seem to care that they didn't predict 2007-8 which as an outsider seems totally bizarre. If their models don't have predictive value, what is the point of any of their work?

Criticising mainstream macroeconomists for the double failure of not seeing 2007-8 coming and also not improving their models after it happened seems completely justified to me. Any macroeconomist who has given up trying to create predictive models because 'people are harder to model than molecules' should be sacked.

Edited February 26, 2019 by Dorkins

[scottbeard]

14 minutes ago, Dorkins said:

Okay, but mainstream macroeconomics totally failed to see 2007-8 coming. That was not a normative question of 'what is the best policy', it was a modelling question along the lines of 'given these macroeconomic conditions and this set of government and central bank policies, what will happen next?'

There are schools of macroeconomic thought that claim to have predicted 2007-8 e.g. followers of Minsky like Steve Keen or the Austrians, but they are on the fringes. Mainstream macroeconomists don't seem to care that they didn't predict 2007-8 which as an outsider seems totally bizarre. If their models don't have predictive value, what is the point of any of their work?

Criticising mainstream macroeconomists for the double failure of not seeing 2007-8 coming and also not improving their models after it happened seems completely justified to me. Any macroeconomist who has given up trying to create predictive models because 'people are harder to model than molecules' should be sacked.

Mainstream modeling did not see 2007-8 coming - I agree, and I think criticizing the modeling for that reason is totally valid. It was a big mistake.

However, tossing out the whole of Economics for that reason would not be fair.  It would be like saying weather forecasting has no predictive value whatsoever, just because Michael Fish got one hurricane warning wrong in 1987.

Do you have a link/evidence to where economists "don't seem to care" or "have not improved their modeling"?  I don't work in economics directly any more, but the financial modeling I do work with is constantly updated (about quarterly) as new data and understanding becomes available - and I can't understand why anyone wouldn't want to improve a model following 2007-8.

[tomandlu]

18 minutes ago, scottbeard said:

 I can't understand why anyone wouldn't want to improve a model following 2007-8.

To give my probably very poor recap of some of the salient points from the book...

1. Most of the assumptions underlying neo-classical economics (NCE) are contradicted by both logic, common-sense, and empirical data
2. NCE depends on the notion of 'rational man' - i.e. that whilst individuals may make bad choices, these errors are flattened in aggregate. Needless to say, most economics books don't even like mentioning the b-word (bubble), or that very few decisions can match the meaning of 'rational' used by NCE.
3. NCE also ignores too many critical issues, such as the environment or an upper limit on resources. Fish stocks, annoyingly, seem immune to the logic of supply and demand.
4. The notion of equal information is not a useful generalisation, as will be obvious if one considers used-car salesmen, a bank, or Facebook for more than five minutes. This may explain why discussions of power-imbalances are almost entirely absent from the standard NCE textbooks.
5. Current assumptions that NCE is somehow 'fair' - i.e. it affords everyone the same opportunity - does not survive even the simplest modelling - under our current system, the rich do get richer, ability and effort be damned, and you can demonstrate this with a couple of lines of code.
6. Better models are available. They will not give us the reassuring (but wrong) predictions of NCE, but they do give us a better picture of potential risks and outcomes (much like chaotic modelling for the weather).
7. Given that NCE cannot, on its own, generate fair or sustainable outcomes, the teaching of economics (and indeed most university subjects) would benefit from a wider curriculum, involving ethics, biology, and so on.

[Riedquat]

51 minutes ago, scottbeard said:

Mainstream modeling did not see 2007-8 coming - I agree, and I think criticizing the modeling for that reason is totally valid. It was a big mistake.

However, tossing out the whole of Economics for that reason would not be fair.  It would be like saying weather forecasting has no predictive value whatsoever, just because Michael Fish got one hurricane warning wrong in 1987..

Have there been any studies into testing the validity of economics? Probably too many factors to make it easily possible.

Clearly there are some useful generalisations that can be made about different possibilities - you can probably be reasonably confident about what happens at the extremes, at least some of the time, such as setting the tax rate to 0% or 100%. That at least defines a space you can work in. The question is how much further can you go from that with better than picking a value at random.

Economics suffers from an expectation that you can be certain (sciences do as well though to be fair), people aren't good at accepting probabilities as answers and that even answers with a high degree of confidence will be wrong sometimes. If you could travel back in time to re-run the same series of events for real you'd not get the same result every time but some might occur more often than others. As you suggest this doesn't make it useless, just not definitive, yet people like certainty that can never exist.

Quote

- and I can't understand why anyone wouldn't want to improve a model following 2007-8.

Of course not but any model of anything will work better in some cases than others - it may well be that the model that would do better there would have serious shortcomings elsewhere. Of course then people will want to try to combine the successful aspects of both but I suspect eventually a wall will be hit (if it hasn't already) where the best any model can do is comparable to predicting the outcome of a roll of the dice - you've got 5 chances in 6 of not getting a 6, but if a 6 comes up it's not a sign of shortcomings of the model.

Edited February 26, 2019 by Riedquat

[tomandlu]

BTW, the most vicious review of the book is by an economist called Chris Auld:

https://chrisaulddotcom.wordpress.com/2011/09/29/a-review-of-economyths-ten-ways-economics-gets-it-wrong-by-david-orrell/

The fun bit is the comments... There is some suggestion that Auld removed some particular pertinent (and damning to him) comments, pointing out where he has either misrepresented the book, or totally missed the point, but TBH I can't see much difference between the live site and the wayback version... your mileage may vary.

[zugzwang]

4 hours ago, scottbeard said:

Mainstream modeling did not see 2007-8 coming - I agree, and I think criticizing the modeling for that reason is totally valid. It was a big mistake.

However, tossing out the whole of Economics for that reason would not be fair.  It would be like saying weather forecasting has no predictive value whatsoever, just because Michael Fish got one hurricane warning wrong in 1987.

Do you have a link/evidence to where economists "don't seem to care" or "have not improved their modeling"?  I don't work in economics directly any more, but the financial modeling I do work with is constantly updated (about quarterly) as new data and understanding becomes available - and I can't understand why anyone wouldn't want to improve a model following 2007-8.

Chote's record of wild misses at the OBR makes Michael Fish look positively clairvoyant!

Is he getting better? How can you tell?! ?

 

 

[tomandlu]

1 hour ago, zugzwang said:

Chote's record of wild misses at the OBR makes Michael Fish look positively clairvoyant!

Is he getting better? How can you tell?!

And to be fair, meteorology was one of the first disciplines to embrace chaos theory as an explanation, if not a solution, as to why the predictions were so violently and unpredictably inaccurate.

[zugzwang]

1 hour ago, tomandlu said:

And to be fair, meteorology was one of the first disciplines to embrace chaos theory as an explanation, if not a solution, as to why the predictions were so violently and unpredictably inaccurate.

Ed Lorenz and his famous Royal McBee. But world + dog caught on to the notion of deterministic chaos in the 1960s and 70s: Astronomers, physicists, biologists, engineers, ecologists.

Everyone it seems except the damn fool economists!

 

[tomandlu]

9 hours ago, zugzwang said:

Everyone it seems except the damn fool economists!

Theories as to why on the back of a postcard, please...

My personal theory is that the other disciplines deal with externals, whereas economics is a human construct, so the assumption was that it should be highly quantifiable - and, more importantly, controllable.

In addition, whilst the other disciplines have strong, empirical, and therefore reassuring, fundamental foundations (physics, essentially) - economics is emergent by its very nature.

[BorrowToLeech]

17 minutes ago, tomandlu said:

Theories as to why on the back of a postcard, please...

My personal theory is that the other disciplines deal with externals, whereas economics is a human construct, so the assumption was that it should be highly quantifiable - and, more importantly, controllable.

In addition, whilst the other disciplines have strong, empirical, and therefore reassuring, fundamental foundations (physics, essentially) - economics is emergent by its very nature.

Economics predicts its own failure. If economists are economically rational, they will produce the theories that are the most profitable, not the most accurate.

 

[zugzwang]

31 minutes ago, tomandlu said:

Theories as to why on the back of a postcard, please...

My personal theory is that the other disciplines deal with externals, whereas economics is a human construct, so the assumption was that it should be highly quantifiable - and, more importantly, controllable.

In addition, whilst the other disciplines have strong, empirical, and therefore reassuring, fundamental foundations (physics, essentially) - economics is emergent by its very nature.

It's the absence of conserved quantities in Economics that makes it more like natural history than science. Emmy Noether taught us that conservation laws follow from invariance principles (symmetry operations) and Eugene Wigner pointed out that mathematics is effective in physics only because of the four local space-time symmetries (translational, time-translational, rotational and Galilean). These four invariances are the basis of mathematical law in Classical Mechanics, Quantum Mechanics and General Relativity. Without them it would be impossible to isolate the system we wish to investigate from its environment; mathematical law might well exist in principle but the system would be too complex for us to identify it in practice. This is pretty much where we stand with Economics, where money as credit is being continuously created, multiplied by leverage and destroyed in an ad-hoc and uncoordinated fashion. As a point of interest, Einstein very nearly employed the term 'Invariance Theory' rather than Relativity.

[Riedquat]

1 hour ago, tomandlu said:

Theories as to why on the back of a postcard, please...

My personal theory is that the other disciplines deal with externals, whereas economics is a human construct, so the assumption was that it should be highly quantifiable - and, more importantly, controllable.

In addition, whilst the other disciplines have strong, empirical, and therefore reassuring, fundamental foundations (physics, essentially) - economics is emergent by its very nature.

That makes the assumption that we create systems that we have control over. Economics (or at least the economic systems we live with) is a great example of one that we really don't. Some limited pushing, with not entirely predictable outcomes, is the best you can do. A good example is how easy it is to effectively end up in (the equivalent of) arms races whether you like it or not - businesses have to behave in certain ways to survive, which aren't necessarily the ways enlightened self interest would dictate but no-one can afford to engage in rational behaviour, or at least what's rational behaviour beyond the very limited constraint of surviving within this system.

Certainly any studies on economics which don't consider, as a very significant part, a study of human behaviour, both individual and en mass, is on a hiding to nothing.

[crouch]

Most people involved in economics are not involved in macroeconomics or forecasting but in much smaller areas. In these areas economics can be useful. For example the work by Foster and Beesley on cost benefit analysis of the Victoria Line was both seminal and useful.

Economic forecasts work by the continuance of trends but one off events like GFC1 are not, by definition part of a trend; they are exceptions to it. 

There are also different schools. The Austrian school which is based on methodological individualism takes an approach based on individual motivation. In this school the idea of forecasting is ridiculous. This is what Hayek, an Austrian, said about forecasting:

 "such complex phenomena as the market, which depend on the actions of many individuals, all the circumstances which will determine the outcome of a process … will hardly ever be fully known or measurable."

The mere fact that there are different schools of economics tells you an awful lot.

[zilly]

20 hours ago, Dorkins said:

 

There are schools of macroeconomic thought that claim to have predicted 2007-8 e.g. followers of Minsky like Steve Keen or the Austrians, but they are on the fringes. Mainstream macroeconomists don't seem to care that they didn't predict 2007-8 which as an outsider seems totally bizarre. If their models don't have predictive value, what is the point of any of their work?

 

Doesn't really seem to me to be that bizarre - they weren't held accountable for their malfeasance, were allowed to claim said malfeasance was due to 'mistakes' (thus redefining in the process what a 'mistake' is, it would seem) - and hence, like a small child which has been allowed to get away with too much, they shrugged their shoulders and got back to doing what they did - in a turbocharged version, having been handed trillions in newly-created CB cash (austerity for the rest of us, natch) with which to play.

Reward for failure generally doesn't engender introspection as to the nature of that failure (which were - of course - merely 'mistakes')....

 

Edited February 27, 2019 by zilly

[Riedquat]

57 minutes ago, crouch said:

Most people involved in economics are not involved in macroeconomics or forecasting but in much smaller areas. In these areas economics can be useful. For example the work by Foster and Beesley on cost benefit analysis of the Victoria Line was both seminal and useful.

Even on those matters I'm somewhat sceptical because both costs and benefits are often both very important and subjective. As long as the limitations of such analysis are thoroughly understood that's OK (you still need a firm grip of the financial costs and benefits) but there's a tendency to come up with excuses to not think any further.

[crouch]

Just now, Riedquat said:

Even on those matters I'm somewhat sceptical because both costs and benefits are often both very important and subjective. As long as the limitations of such analysis are thoroughly understood that's OK (you still need a firm grip of the financial costs and benefits) but there's a tendency to come up with excuses to not think any further.

Yes they are, you're right. But that doesn't mean that anaysis isn't useful.

On the Victoria Line an accounting definition would most likely see it as a heavy loser.

However, CB analysis seeks to evaluate the effects of: taking traffic off the roads; a reduction in pollution leading to lower health costs.... Now you're right there may be items going the other way which are not recognised but, at the very least, it does say that the accounting model has severe limitations when considering public works projects like this.

[Riedquat]

I'm not trying to say that such analysis isn't useful. I think it comes back to an oft-repeated view, that the subjective, personal, even sentimental tells you what's desirable but you need a firm grip of the facts and figures to work out how to achieve it, and which means, within those subjective constraints, are the best ways of achieving that. Therefore the ability to perform such analysis (and I'm glad it took in a wider picture than direct financial costs alone) is an incredibly useful tool for achieving what's desirable. But it doesn't itself tell you what's desirable.

It's also subject to a lot of uncertainties due to the uncertainty of human behaviour; they can be reduced but not eliminated, and whilst such localised considerations might be less prone to finger-waving-in-the-air than macroeconomics they still suffer from some of the same problems. You can have guesstimates about how many people will change to the Tube from the car, but only that. Of course there's no alternative to living with that.

[crouch]

16 minutes ago, Riedquat said:

I'm not trying to say that such analysis isn't useful. I think it comes back to an oft-repeated view, that the subjective, personal, even sentimental tells you what's desirable but you need a firm grip of the facts and figures to work out how to achieve it, and which means, within those subjective constraints, are the best ways of achieving that. Therefore the ability to perform such analysis (and I'm glad it took in a wider picture than direct financial costs alone) is an incredibly useful tool for achieving what's desirable. But it doesn't itself tell you what's desirable.

It's also subject to a lot of uncertainties due to the uncertainty of human behaviour; they can be reduced but not eliminated, and whilst such localised considerations might be less prone to finger-waving-in-the-air than macroeconomics they still suffer from some of the same problems. You can have guesstimates about how many people will change to the Tube from the car, but only that. Of course there's no alternative to living with that.

Indeed. But the context is different from a private project.

A private project is meant to measure private return and the base parameters are those legally contracted costs and revenues that are necessary to the project.

A public project is meant to evaluate public return which is a different concept and returns different things even than a private project.

And you are correct that it doesn't tell you what is desirable. Desirability is what is attractive, useful or necessary. Attractiveness is difficult to measure; useful is a matter of degree and necessary may be a matter of opinion.

But in the appropriate context it can be a useful technique. In the case of the Victoria Line you might argue that it is simply a way of justifying the project rather than evaluating it and that a more legitimate use would be in comparing projects but this may be difficult.

Edited February 27, 2019 by crouch

[zugzwang]

1 hour ago, zilly said:

Doesn't really seem to me to be that bizarre - they weren't held accountable for their malfeasance, were allowed to claim said malfeasance was due to 'mistakes' (thus redefining in the process what a 'mistake' is, it would seem) - and hence, like a small child which has been allowed to get away with too much, they shrugged their shoulders and got back to doing what they did - in a turbocharged version, having been handed trillions in newly-created CB cash (austerity for the rest of us, natch) with which to play.

Reward for failure generally doesn't engender introspection as to the nature of that failure (which were - of course - merely 'mistakes')....

 

Former Fed Chairman Benny 'The Bubble' Bernanke repeatedly upbraided hedge funds and and over-leveraged speculators for their part in the GFC, and then joined Citadel twelve months after retiring.

What a schmuck.

 

Edited February 27, 2019 by zugzwang

[scottbeard]

10 hours ago, Riedquat said:

That makes the assumption that we create systems that we have control over. Economics (or at least the economic systems we live with) is a great example of one that we really don't. Some limited pushing, with not entirely predictable outcomes, is the best you can do. 

The interesting thing though is that, whilst we don't have total control over the economic systems we create, we CAN influence them.  They are constantly changing.

Those lucky mathematicians can work safe in the knowledge that 2  + 2 always = 4, and there is nothing anyone can do to change it.  The reason we still use Maths first described by the ancient Greeks like Pythagoras or Euclid is that it's still just as true today.  Alas Economists work in a world which is constantly evolving - where suddenly nuclear power is discovered, or women start working in vast numbers, or the internet is invented, or interest rates fall to zero for decades - and all their past data doesn't work any more.  I think Economics is harder to get right than natural sciences.

[Riedquat]

Oh, definitely. It's easy to laugh at economics, and it does seem to have issues with its dogmatic gospel truths, but there are group behaviours under certain circumstances that are more likely than others and those can be used as the basis for models, under certain assumptions. When those assumptions are broken, which the type of random events you describe do (of which the interest rates is probably the least random) they'll inevitably fail. Since those things probably crop up every decade or two that's as far as it's worth trying to guess.