PDA

View Full Version : astrology and economics



profloater
2013-Mar-19, 12:55 PM
astrology and economics

This is a subversive thread about correlation and causation. Or lack of causation. Astrology is roundly abused for lack of causation but has a rich vocabulary of archetypes and a cyclic basis rooted in undeniable observable events. Indeed the vocabulary of greek myths and their archetypes is wider by far then psychological profiles accepted in scientific circles but the baby there gets thrown out with the bathwater.

Economics also has a vocabulary or jargon to impress and exclude outsiders but is accepted as a tool of political strategy and tactics very much as astrology used be say 400 years ago. But I suggest the successes of economics in a causal understanding are just as limited as the tides and the moon, or the fur trapping cycle and weather with Jupiter. That is there is no rational extension to other areas of life.

You can analyse the causes of the great depression, say, or the world wars in terms of economic theory or astrological cycles with about equal amounts of impressive coincidences. We can always find patterns in the data.

Early economic theory was based on the risible notion that all players are rational and in command of all the relevant market facts. Even then a rational person would point out that every deal in the world has a buyer and a seller, and both feel they are getting a good deal and they cannot both be right. The market facts can never anticipate future events and the intended or unintended consequences for each deal. Unlike astrology which assumes cycles, early economics assumed long term trends based on economic theory but boom and bust is the factual history. Quaint that the boom and bust cycle correlates with the mythical fear and greed planets Jupiter and Saturn, of course that's a coincidence.

HenrikOlsen
2013-Mar-19, 01:14 PM
Even then a rational person would point out that every deal in the world has a buyer and a seller, and both feel they are getting a good deal and they cannot both be right.
I question this statement as it's apparently based on an axiom that things have intrinsic value.

Different people have different current needs, which means they have different valuation of different wares.

That the wares have different value for different people makes it entirely possible for an exchange to be a good deal for both.


Simplest example, I picked a flower in the forest, you picked an apple.
I don't like roses but like apples, you don't like apples but like roses.
We swap.
This was a good deal for both of us.

Your axiom is wrong for the real world.


Quaint that the boom and bust cycle correlates with the mythical fear and greed planets Jupiter and Saturn, of course that's a coincidence.Post the data you are basing that claim on (in a thread in ATM) and let us run the numbers for real.

profloater
2013-Mar-19, 01:27 PM
Yes you are right about fair trades but I was thinking of the stock market mainly. I did not choose ATM because I do not want to defend astrology as such, just to point out that that correlations are there. If you like I can certainly post some illustrations to play with.

grapes
2013-Mar-19, 03:05 PM
Even then a rational person would point out that every deal in the world has a buyer and a seller, and both feel they are getting a good deal and they cannot both be right.
No, they can both be right, even if you limit the situation to the stock market mainly.

Later, they may feel differently. :)

I did not choose ATM because I do not want to defend astrology as such, just to point out that that correlations are there. If you like I can certainly post some illustrations to play with.
Doesn't matter. If you pick twenty scenarios, you're going to have a fit for one of them at the 5% level. The problem is starting with a model, and using it to predict--that doesn't happen.

The danger with such economic models is they tend to feedback on themselves. :)

profloater
2013-Mar-19, 03:13 PM
On the way to some actual figures may I mention the Kondratiev cycles? He was an economist (not an astrologer) who claimed to see about 60 year business cycles in history. This was in the 1920s. He saw them in three phases: development, stagnation, collapse, approximately equal. Later fans see four subcycles but recently Korateyev, another economist used a mathematical approach to confirm three subcycles from economic data.

What would an astrologer say? The business cycle is fear and greed as I mentioned. The saturn Jupiter conjunctions as seen from earth are about 20 years apart, (so three of them is one K cycle) The conjunctions of the 20th century were late in 1901, autumn 1921, autumn 1940, early 1960 and early 1981, (again in 2001)

Frankly these do not directly fit the data but loosely they do. (Kondratiev reckoned his cycle restarted in 1896) 1921 was a recession after the first world war, 1940 was the beginning of the second world war the sixties were the sixties and 1981 was a dow jones low.

Point is you can side with astrologers or economists, correlation hunting.

Of course many modern economists do not accept K cycles, but some still do.

grapes
2013-Mar-19, 04:02 PM
Point is you can side with astrologers or economists, correlation hunting.
As long as you stick with economic cycles, in this thread. No astrology, it's been done before.

profloater
2013-Mar-19, 04:21 PM
My subversive point is that economists get nobel prizes and help control the nations of the world because they dress up their correlations as theories.

Noclevername
2013-Mar-19, 05:11 PM
http://www.smbc-comics.com/index.php?db=comics&id=2845#comic

COMIC:

Guy 1: "Hey, you know how we sit around every night grousing about the way nobody agrees with us about how society works?"

Guy 2: "Yep."

Guy 1: "What if we mathematized that?"

CAPTION: BEFORE ECONOMICS.

Alt text:
G1: "The humans aren't fitting my equation."

G2: "Replace them!"

Mild warning, this one strip is SFW, others are not.

Gillianren
2013-Mar-19, 05:27 PM
I question this statement as it's apparently based on an axiom that things have intrinsic value.

A very good friend makes his living selling things, things he's made himself. He's extremely happy at it. Possibly one of the happiest people I know. (Though he wishes his daughter and his girlfriend would get along; I can't help him there.) Yes, he gets your money, but he also gets your pleasure at finding The Thing You Wanted, the pleasure of making the things, the experience of selling them. His job is fun for him, and he always says that, while he could charge more if people insist, the amount he charges is what's right for him. His wares are ridiculously cheap, given the workmanship and materials.

profloater
2013-Mar-19, 05:44 PM
being a designer maker myself, I can relate to that very well. There is a great pleasure in making things (and repairing things) even when the "economics" appear to be against you. I used to know a skilled woodworker whose rates were really low, but he said he like to be able to take his own time to get it right, and of course he was always in demand. My aim here is at poking fun at fancy economic theories, just for fun. Take monetarism for example, where we are now I believe. You control the behaviour by controlling the money supply, it's positive feedback, which can be shown to lead to oscillation.

grapes
2013-Mar-19, 05:56 PM
My subversive point is that economists get nobel prizes and help control the nations of the world because they dress up their correlations as theories.
They get Nobel Memorial Prizes. :)

starcanuck64
2013-Mar-19, 10:38 PM
When you consider that significant areas of modern commerce exist in a black box environment, such as derivatives, then it is possible to compare economics to astrology. Factually they may have a similar basis, involving peoples acceptance of a perceived reality rather than an established one.

Think of Allan Greenspan being an economic wizard for decades that could seemingly not get anything wrong, suddenly being proved fundamentally wrong when long term trends finally led to serious economic declines.

HenrikOlsen
2013-Mar-19, 10:54 PM
Think of Allan Greenspan being an economic wizard for decades that could seemingly not get anything wrong, suddenly being proved fundamentally wrong when long term trends finally led to serious economic declines.
Or think of it another way: for decades, investment banking was a cult of personality where investors believed in what Greenspan said they'd do, and therefore did it.
With him gone, no longer was there a single person to whom everyone looked for direction, so the basic lack of predictability in a market driven by irrational players with incomplete knowledge started peeking through the cracks in the walls of the castle of mirrors people had built on a foundation of faith in him knowing what was going on.

Hlafordlaes
2013-Mar-19, 11:12 PM
When you consider that significant areas of modern commerce exist in a black box environment, such as derivatives, then it is possible to compare economics to astrology. Factually they may have a similar basis, involving peoples acceptance of a perceived reality rather than an established one.

Think of Allan Greenspan being an economic wizard for decades that could seemingly not get anything wrong, suddenly being proved fundamentally wrong when long term trends finally led to serious economic declines.

Nail hit with hammer on BAUT\CQ.

starcanuck64
2013-Mar-19, 11:14 PM
Or think of it another way: for decades, investment banking was a cult of personality where investors believed in what Greenspan said they'd do, and therefore did it.
With him gone, no longer was there a single person to whom everyone looked for direction, so the basic lack of predictability in a market driven by irrational players with incomplete knowledge started peeking through the cracks in the walls of the castle of mirrors people had built on a foundation of faith in him knowing what was going on.

Greenspan effectively posed the hypothesis that no matter what players did in the market, forces within the market would always bring stability and growth in the long term and thus justifying the removal of regulation that was created to avoid the kind of behavior that led to earlier collapses.

I think what has occurred in recent decades has been a continual decline in overall economic stability as controls have been removed while Greenspan and others did a lot of handwaving from the sidelines. In my opinion, they can be thought of more as magicians than experts, kind of like astrologers.

publius
2013-Mar-19, 11:28 PM
The problem with economics, as a discipline, is it suffers from physics envy. Economics is always going to be a soft science, just like psycho/sociology, and indeed, perhaps ought to consider itself a subset of that, and they themselves suffer from a bit of physics envy themselves although that is more of a lot of statistical razzle-dazzle on the data from various studies where they get caught up in the math of that and somehow think that means something about the human mind subjects of those studies. Economics goes a little further than that, imagining the humans can be described by equations themselves.

Human beings aren't electrons, and they are never going to be, and there is no set of equations that is going to describe and predict human behavior. You can write all sorts of equations and construct models but no set of humans is ever going to follow that model, because they ain't electrons.

Economics does have a "physics" like component that depends on objective reality as it were, but the subjective part of the human minds that interact it with it will always be non-objective.

The notion of "value" has no objective definition and thus a hard science treatment is impossible. I remember posting something a good while back in the Black Monday thread, a piece by some economics big shot who used to work at the Fed, then moved on, who pointed out just how poor economic models were and the problems of trying to draw conclusions from existing economic data.

His point was how various central planner types demand to know what the result of policy A will be, or what should the policy be in order to achieve outcome B. An honest economists would say "I can't answer that question with certainty". But if they did that, the various modelers would be out of a job, and the central planner types and their political masters just don't want to hear that, and thus will listen to anyone who claims to be able to answer the question.

So in that sense it might be like some king of old consulting his astrologers and soothsayers. :)

starcanuck64
2013-Mar-19, 11:44 PM
It's very hard to accept that economics will ever be able to accurately predict outcome when very influential players in the market do the most they can to conceal their actions. With derivatives which is still largely unregulated and even un-monitored few people understand what's going on even though trillions of dollars are invested. I think the 1980s term Voodoo economics is still applicable today.

If you look at this story, it's hard to understand even what's already occurred let alone predict where events will lead.

http://business.financialpost.com/2013/03/15/jpmorgan_jamie-dimon_london-whale/


Bloomberg News first reported on April 5 that the portfolio built by U.K. trader Bruno Iksil, known as the London Whale, had built an illiquid book of derivatives so large that it was distorting credit indexes. The Federal Reserve and Office of the Comptroller of the Currency sought additional information about the trades following media reports.

grapes
2013-Mar-19, 11:45 PM
Mild warning, this one strip is SFW, others are not.
Engineers "doing" economics:
http://www.smbc-comics.com/index.php?db=comics&id=2855#comic

profloater
2013-Mar-20, 09:21 AM
consider the chartists: I mean those who invest by watching the trend lines. They are in direct opposition to the fundamentalists: I mean those who analyse P/E ratios, lots of ratios. Indeed the one bunch hope to gain from the mistakes made by the other. The one is assuming they can predict from previous behaviour and the other assumes that logic must prevail. We have seen that they both get it wrong. If economists would admit that the market is irrational and chaotic, that would be honest, but they pretend they have the theory that will work and give the right result. But of course there are overall psychological swings in a democracy. (Other systems have different dynamics) There is only one true logic operating in my view, the feeling that it's time to give the other lot a go. This must be, if true, a distinctly depressing thought for an honest politician. It's a cyclic thing.

caveman1917
2013-Mar-20, 09:35 AM
When you consider that significant areas of modern commerce exist in a black box environment, such as derivatives, then it is possible to compare economics to astrology.

How are derivatives a black box environment? I'd say they're less of a black box environment than the stock market itself, it just allows you to invest in different moments of a stock's time series. Take for example investing in options (a derivative) compared to investing in stocks themselves. In as far as the efficient market hypothesis holds the conditional mean of a stock equals its unconditional mean (basically you can't know if the stock will be higher or lower tomorrow than it is today). However with options you invest in the conditional variance which is not the same as its unconditional variance (you can know if it will make a big move tomorrow or a small move - even if you don't know what direction that move will be in). Now that options market itself is then again quite efficient, but it appears to be less so than the direct stock market. Or perhaps betting would be a better term than investing, but still.

profloater
2013-Mar-20, 09:51 AM
Derivatives would only be efficient if you can bet both ways and still make a profit whatever happens. They are surely in themselves pure gambling. As I understand it, the proper purpose of options is to bet the other way when buying or selling actual stocks in order to reduce losses if you have made a mistake. Of course derivatives are tools of tactics rather than economic theory. They are a practical tool recognising that the market is irrational. What I am pointing at is the predictions made by economists both of cycles and the effect of government policy which are held to be "scientific" or "rational" when they are about the same as saying "sell when Mercury is retrograde"; that is right sometimes and wrong sometimes.

SeanF
2013-Mar-20, 01:18 PM
Greenspan effectively posed the hypothesis that no matter what players did in the market, forces within the market would always bring stability and growth in the long term and thus justifying the removal of regulation that was created to avoid the kind of behavior that led to earlier collapses.

I think what has occurred in recent decades has been a continual decline in overall economic stability as controls have been removed while Greenspan and others did a lot of handwaving from the sidelines. In my opinion, they can be thought of more as magicians than experts, kind of like astrologers.
Wait a second -- are you equating "economic experts" with magicians/astrologers, and at the same time claiming that our current financial problems are a result of "economic experts" not having enough control over the system?

grapes
2013-Mar-20, 01:57 PM
Wait a second -- are you equating "economic experts" with magicians/astrologers, and at the same time claiming that our current financial problems are a result of "economic experts" not having enough control over the system?
Maybe it's "controls work, but we don't understand the controls"

Seems dubious. Although there is still the analogy to physics...

peteshimmon
2013-Mar-20, 02:00 PM
Thinking of how some people let star forecasts
guide their day, a forecast that specifically
advises on their spending patterns may allow
direct influence on economic activity. A prospect
that may start some treasury people in governments
to start drooling! Dont give them ideas.

:)

Trebuchet
2013-Mar-20, 03:25 PM
A very good friend makes his living selling things, things he's made himself. He's extremely happy at it. Possibly one of the happiest people I know. (Though he wishes his daughter and his girlfriend would get along; I can't help him there.) Yes, he gets your money, but he also gets your pleasure at finding The Thing You Wanted, the pleasure of making the things, the experience of selling them. His job is fun for him, and he always says that, while he could charge more if people insist, the amount he charges is what's right for him. His wares are ridiculously cheap, given the workmanship and materials.

I'm guessing I've both met your friend and seen his stuff, which is pretty awesome.

The difference between astrology and economics is that one is a pseudoscience based on handwaving and mumbo-jumbo, while the other ... hmm, guess there's no difference after all.

profloater
2013-Mar-20, 04:54 PM
I'm guessing I've both met your friend and seen his stuff, which is pretty awesome.

The difference between astrology and economics is that one is a pseudoscience based on handwaving and mumbo-jumbo, while the other ... hmm, guess there's no difference after all.

The similarities are, I am saying, forcing patterns onto correlations to find trends and cycles without any causal relationship beyond crude causal relationships which support the theories, until experience goes against the theories, but somehow the theories survive..

Perikles
2013-Mar-20, 05:26 PM
The similarities are, I am saying, forcing patterns onto correlations to find trends and cycles without any causal relationship beyond crude causal relationships which support the theories, until experience goes against the theories, but somehow the theories survive..I find these similarities quite convincing. In both areas (I typed 'disciplines' but it looked ridiculous) there always seems to be a theory to explain past sequences of events. The theory then fails spectacularly to predict an event, and apologists leap in with a refinement to the theory to close the loophole. This behaviour repeats itself indefinitely.

I forget the detail, but the event which conditioned my attitude to economics most was a report at the time some 4 decades ago that a finals exam for Oxford students reading PPE (E=economics) had a question which went like "Explain why an economic situation of a country involving phenomena W, X, Y and Z simultaneously is actually impossible". Very shortly after that, the UK found itself to be in exactly that 'impossible' dire situation.

Gillianren
2013-Mar-20, 05:32 PM
I'm guessing I've both met your friend and seen his stuff, which is pretty awesome.

You have, yes. One of the reasons his daughter doesn't like his girlfriend is that his girlfriend wants him to quit and get a "real job." Which would make him miserable.

Carla
2013-Mar-20, 05:43 PM
I find these similarities quite convincing. In both areas (I typed 'disciplines' but it looked ridiculous) there always seems to be a theory to explain past sequences of events. The theory then fails spectacularly to predict an event, and apologists leap in with a refinement to the theory to close the loophole. This behaviour repeats itself indefinitely.

If that's how it works, it's just embarrassing. I's no better than physics.

profloater
2013-Mar-20, 07:58 PM
If that's how it works, it's just embarrassing. I's no better than physics.

In that there are dangers in correlation without causation you may be right but I did not say that explicitly.;)

profloater
2013-Mar-20, 08:07 PM
Wait a second -- are you equating "economic experts" with magicians/astrologers, and at the same time claiming that our current financial problems are a result of "economic experts" not having enough control over the system?

Er, yes but...I do not favour experts having control of the system, the idea of control is , I maintain, an illusion, a dangerous illusion.

I am reminded of the hypothetical physical problem of dragging a reluctant dog along by a string. When the string is tied to the collar, you have some control but if you tie the string to the tail it becomes chaotic and may run backwards or bite you. Historical analysis is the collar case, trying to control the future is the tied tail case.

I am not sure how far one can take that humble metaphor, but it is a control theory case study!

caveman1917
2013-Mar-20, 08:44 PM
Derivatives would only be efficient if you can bet both ways and still make a profit whatever happens.

That's impossible. If i had a "product" (set of derivatives) that would guarantee a profit of say 100$, why would i sell it to you for less than 100$? What you describe is probably the most inefficient market imagineable. What would happen is that someone will pay say 10$ for that, but then someone else will offer 20$ and so on until someone offers 99.99$ which becomes the market price of that set of derivatives and the guaranteed profit disappears, or at least becomes less or equal than the transaction costs.


They are surely in themselves pure gambling.

Sure, but no more than "normal" investing. If you buy a stock you're betting that it will go up, if you sell one you're betting that it will go down. It's a bit more complicated than that, but it's mathematically equivalent to a gambling system, you're paying money that will get you a profit depending on the outcome of a future event. I just find it funny that if one is betting on the future first moment of a timeseries it's called "investing" but if you're betting on the future second moment of that timeseries it's called "gambling". Derivatives are no more gambling or more blackbox-like than the primary market.


As I understand it, the proper purpose of options is to bet the other way when buying or selling actual stocks in order to reduce losses if you have made a mistake.

That's how they are explained yes, but the interesting thing is that you don't need options for that. There are other tools that do that just as well that have always been available (they're called guaranteed stop orders). They are also heavily used for short-selling since there is now a law precluding you from doing that directly on the primary market. It does make one wonder as to the usefulness of that law, given that you can just as well short-sell a stock by using options.

publius
2013-Mar-20, 09:36 PM
SPeaking of "control", I'll give you an example of how things just aren't as cut and dried as the central planner types think. On another forum, I casually mentioned something about the Fed "controlling" interest rates, via their control of the money supply. Somebody replied, "Oh really" and showed my an interesting little Fed graph of 3-month T-bill rates (a benchmark for short term rates) vs the Fed Funds rate. They correlated perfectly in the aggregate big picture, but guess what? The T-bill rate moved *before* the Fed Funds rate. T-bill rates are set at auction and on the secondary market, and they were moving before the interbank rate. His claim was the market moves the Fed, not the other way around.

ANd he had a good point. I found a long winded paper on this very thing that looked at all that closely. It involved a lot oh heavy statistical analysis (Granger causation and all that good stuff). The conclusion was that the T-bill rate is moved by what the market *thinks* the Fed is doing, not what it really is doing. There were times when the market erroneously thought the Fed had made a change when it hadn't and the T-bill rate moved according to the false information.

And the Fed Funds rate is even more complicated and not much could be said about it, other than sometimes the Fed moves it, and sometimes it moves the Fed. While the big picture macro correlation with T-bill rates is good, there's a lot of complexity in the details.

So thus is perhaps better to say that the Fed's control is more "open mouth operations" rather than open market operations. The market believes the Fed has great power, and thus acts when it thinks the Fed has exercised that power.

What would be really interesting is a stealth central bank, one the market didn't know about it, making open market money supply operations that no one knew about. That would change the reserve supply, but would it even make a dent in what the market actually did? (Until the massive QE operations that have tripled the monetary base since the start of the crisis, the magnitude of OMOs was actually quite small relative to the base).

It's quite complicated, depending on human behavior, and not some simple equations about money supply and interest rates and all that. If the market players were electrons and not human beings, such models would be good. But they ain't and so they aren't. If the models work to any extent, it's probably more because people believe they work and thus act accordingly. :)

caveman1917
2013-Mar-20, 10:28 PM
SPeaking of "control", I'll give you an example of how things just aren't as cut and dried as the central planner types think. On another forum, I casually mentioned something about the Fed "controlling" interest rates, via their control of the money supply. Somebody replied, "Oh really" and showed my an interesting little Fed graph of 3-month T-bill rates (a benchmark for short term rates) vs the Fed Funds rate. They correlated perfectly in the aggregate big picture, but guess what? The T-bill rate moved *before* the Fed Funds rate. T-bill rates are set at auction and on the secondary market, and they were moving before the interbank rate. His claim was the market moves the Fed, not the other way around.

ANd he had a good point. I found a long winded paper on this very thing that looked at all that closely. It involved a lot oh heavy statistical analysis (Granger causation and all that good stuff). The conclusion was that the T-bill rate is moved by what the market *thinks* the Fed is doing, not what it really is doing. There were times when the market erroneously thought the Fed had made a change when it hadn't and the T-bill rate moved according to the false information.

It's the same thing with market inefficiencies around central bank interventions. I also read a paper taking a deeper look into the observation that the market shows noticeably less efficiency on the days of central bank interventions and it found the same conclusion. The inefficiency comes before the intervention, not afterwards. It is again the market's expectation of the intervention that is causing this, not the intervention itself.


If the models work to any extent, it's probably more because people believe they work and thus act accordingly.

That's a nice counterargument to so-called technical analysis models. If it really were true that one could deduce from those models that say a stock will go up tomorrow, then everyone would buy that stock making it go up in price immediately up until the point where you wouldn't expect it to go up even more tomorrow.

profloater
2013-Mar-20, 11:20 PM
the self fulfilling prophesy is a real effect and some of the players are not above manipulation by sheer size. Especially knowing that computer based decisions will watch for volume of trade and pile in, so causing a rise which can be sold. However even the fifty of more economic theories with smart sounding names do not include the effect of prophecy as far as I know. I believe it was Leibnitz that started the ball rolling for chaos theory and that theory shows there is an envelope of possibility but tiny changes can put you anywhere after quite a short time. The weather is chaotic but you can find trends, but these have simple seasonal drivers. Obviously the financial globe has many drivers (including the weather) and we do not know all the envelope of possibilites. The clause "uncharted waters" keeps popping up in the financial news.

starcanuck64
2013-Mar-21, 01:15 AM
How are derivatives a black box environment? I'd say they're less of a black box environment than the stock market itself, it just allows you to invest in different moments of a stock's time series. Take for example investing in options (a derivative) compared to investing in stocks themselves. In as far as the efficient market hypothesis holds the conditional mean of a stock equals its unconditional mean (basically you can't know if the stock will be higher or lower tomorrow than it is today). However with options you invest in the conditional variance which is not the same as its unconditional variance (you can know if it will make a big move tomorrow or a small move - even if you don't know what direction that move will be in). Now that options market itself is then again quite efficient, but it appears to be less so than the direct stock market. Or perhaps betting would be a better term than investing, but still.

With derivatives you're essentially selling insurance on other stocks performance and using very complex mathematical models to do so. Often the information is incomplete and sometimes criminally so although it's hard to enforce laws in an area where there are few regulations and very little oversight. As we've seen trillions of dollars can seemingly quickly vanish. There's a lot of money to be made for a few people in the right spot at the expense of destabilizing the entire market.

PBS did an excellent program on the issue;

http://www.pbs.org/wgbh/pages/frontline/warning/interviews/born.html


It was like my worst nightmare coming true. I had had enormous concerns about the over-the-counter derivatives [OTC] market, including credit default swaps, for a number of years. The market was totally opaque; we now call it the dark market. So nobody really knew what was going on in the market.

If you have over 10 times the gross national product of the entire world locked up in a sector that is essentially opaque, how can there be any realistic forecasting or regulation?


We had no regulation. No federal or state public official had any idea what was going on in those markets, so enormous leverage was permitted, enormous borrowing. There was also little or no capital being put up as collateral for the transactions. All the players in the marketplace were participants and counterparties to one another's contracts. This market had gotten to be over $680 trillion in notional value as of June 2008 when it topped up. I think that was the peak. And that is an enormous market. That's more than 10 times the gross national product of all the countries in the world.

HenrikOlsen
2013-Mar-21, 07:49 AM
All these things are not really news, the market is already known to be driven by expectations, it's built in. it's a lot of people gambling on what the rest will do in the future, with some hypotheses about what additional influences exist.
It's axiomatic in the models that there are external influences, but their effects are just as unknown as anything else because the link is through expectations rather than causal.

Ivan Viehoff
2013-Mar-21, 09:52 AM
Early economic theory was based on the risible notion that all players are rational and in command of all the relevant market facts. Even then a rational person would point out that every deal in the world has a buyer and a seller, and both feel they are getting a good deal and they cannot both be right. The market facts can never anticipate future events and the intended or unintended consequences for each deal. Unlike astrology which assumes cycles, early economics assumed long term trends based on economic theory but boom and bust is the factual history. Quaint that the boom and bust cycle correlates with the mythical fear and greed planets Jupiter and Saturn, of course that's a coincidence.
I'm an economist, I do it for a living. Not macro-economics though, business economics.

You have a misunderstanding of the role of the "rational agent" assumption in economics. A lot of good modern, testable economics is still based on it, and there is a good reason for that. It isn't always a risible assumption. Even when it is a dodgy assumption, an alternative assumption may make the issue intractable, but at least the rational agent model is standard against which you can try and adjust for the effect of irrationality.

The reason it is often a good assumption is that in many aggregate situations, individual irrationalities cancel out on average, so it is a very good working assumption for the prediction of the behaviour of a crowd. The clever bit is to recognise when it isn't an adequate assumption. My main field of work is transport, and when trying to forecast passenger behaviour to a change in ticket prices, the rational agent assumption remains a very good one.

It's like the assumption of an ideal gas in physics or chemistry. For many purposes it is a perfectly adequate assumption. The clever bit is to recognise when it isn't an adequate assumption.

Of course economists have always known that rationality was an idealised assumption. They needed to make it to make the situation tractable to analysis at all. You say "early" but actually there had been very little work on measuring the extent of people's irrationality in the days when I was studying economics. The Allais paradox - demonstrating that people's decisions can be biased rather than equally distributed around the rational ideal - goes back a long way. But the intense study in the area, encouraged by Kahneman and Tversky, is pretty recent.

Anyway, so now we know a bit more about irrationality and have measured its effect in certain very simple, and often artificial, situations. If you think there is very much recognition of it in macro-economic models, I'll have to disabuse you of that. Large scale macro-economic models are still largely general equilibrium models because we still don't know how to do anything else. Off-model analysis is done of real-world effects that the models can't do. The main issue in the present crisis is that of liquidity. Still rather abstract and distant from irrationality.

The other big issue in macro-economics which the models fail to reflect is non-clearing markets. For example the governments have set a low interest rate, and there just isn't enough capital around to lend to everyone who wants to borrow at that interest rate. So the capital market isn't clearing. Housing markets and labour markets don't clear either, as there are institutional reasons why the prices don't easily move to clear them. We recognise this to some degree through liquidity constraints, which even themselves are too difficult to put in the macro models, but the broader issue of all sorts of markets that don't clear properly isn't properly reflected in the models either. This issue has been recognised for over 30 years, but we don't know how to fix it.

Perikles
2013-Mar-21, 10:41 AM
Large scale macro-economic models are still largely general equilibrium models because we still don't know how to do anything else. Thanks for the excellent post. Is it a belief amongst economists that general equilibrium models will eventually be replaced?

profloater
2013-Mar-21, 11:06 AM
I also wish to thank you for this post, I admitted to being subversive in the way I presented the causation case. While I can see the smoothing effect of assuming rationality over a large group there are also clearly adverse crowd effects in both physical crowds and groups of investors where panic and indeed oratory can sway groups away from rationality. Arthur Koestler identified this trait of followship. My take on that is that in order to become a cooperative society we evolved strong followship instincts which are often contrary to our individual best interests, and this always tends to surpress rationality. So sitting down to really think things out and ignore your peers is quite unusual behaviour. Thus to assume that a market will be rational is not necessarily a good assumption. I think we have plenty of examples around us today.

Liquidity; in another thread by Publius the central theme of today's crisis is debt. We could argue that the accumulation of debt is both irrationally optimistic and cynically short termist. The addition of helicopter liquidity is controversial. And would I be right in assuming that general equilibrium model of say an ocean would be a flat calm with tides?

Ivan Viehoff
2013-Mar-21, 02:23 PM
Thanks for the excellent post. Is it a belief amongst economists that general equilibrium models will eventually be replaced?
People have been trying and failing to get different kinds of models for a long time. The economists I've spoken to didn't seem very hopeful that there was any prospect of a different kind of model. They rather think that people will just have to be more aware of the shortcomings of the models and use work-arounds to deal with them. Deep financial crises don't turn up often enough for us to have a clear view of how they work in detail, together with the modern features of the world that didn't exist last time there was one.

Gillianren
2013-Mar-21, 04:45 PM
People have been trying and failing to get different kinds of models for a long time. The economists I've spoken to didn't seem very hopeful that there was any prospect of a different kind of model. They rather think that people will just have to be more aware of the shortcomings of the models and use work-arounds to deal with them. Deep financial crises don't turn up often enough for us to have a clear view of how they work in detail, together with the modern features of the world that didn't exist last time there was one.

And really, that's not the sort of thing you want to happen more often, even if it improves the models.

HenrikOlsen
2013-Mar-21, 05:24 PM
As I see the economic world, including the economists, a fundamental problem is that the economy has multiple modes where each mode by itself can be modeled fairly well, with lots of assumptions and approximations, but it's basically modeled as a system of feedback loops with various external knobs and various amplifications and delays (sometimes negative when the feedback is from expectations).

The problem is that even though the delays and amplitudes are fairly stable as long as the economy stays within one mode, which makes it possible to make fairly accurate models with at least some predictive power, if it flips to a different mode all parameters gets reshuffled, both feedback/control strengths and delays.
Which makes any existing model invalid because the economy is operating outside its realm of applicability and therefore it is useless as a predictor until the economy has been it the mode long enough to get empirical data on how things behave.

For instance, Alan Greenspan's model of economics was perfectly correct within it's real of applicability, which was an economy which to a large extent was governed by the belief that Alan Greenspan knew what he was doing. People believed that he knew what he was doing, so they knew what the rational reaction to his actions was, which made them predictable and his actions useful for regulating things.

Once the mode switched and this belief disappeared, the fed's ability to reliably control anything was lost as well.

starcanuck64
2013-Mar-21, 06:18 PM
It seems to me economics is really an attempt to rationalize what's already happened in the hope of maintaining consumer confidence.

publius
2013-Mar-21, 11:30 PM
I was just reading something. Some outfit in Zurich has just published a paper that claims the 60 -70% of all moves in the commodities markets (which is what they studied, I think) are due to the system itself, and not any exogenous information.

Is the culprit "technical analysis" type self-fulfilling prophecy by the traders as Caveman above mentioned? Well, this seems to be the emergent behavior of HFT systems, which we've discussed in Black Monday on and off again. It was only about 30% "endongenous" before the rise of the machines. So, maybe humans themselves would tend to do it, but you need a computer to really go wild with it.

So heck, soon markets may be nothing but computer simulations....

Really, this HFT stuff needs to be stopped. If you're not familiar with, read about some of the analysese of the various "flash crashes" that have occured, where the computers went wild. There's a whole fascinating world of the emergent behavior of these systems that exist at the microsecond level. Anyway, there's several proposals (which aren't going anywhere because there's so much money involved with the HFTers) that would limit things to human reaction times.

caveman1917
2013-Mar-22, 08:00 AM
With derivatives you're essentially selling insurance on other stocks performance and using very complex mathematical models to do so.

The models aren't that complex, you pay me say 1000$ for the right to buy 100 stocks of company X at price Y$ at the end of the week, can't get much clearer than that. The pricing models can get complex, but so are those in the primary market. Ask anyone for their model about what is the "right price" for a stock, it'll be just as complex as their model for the "right price" of a certain option on that stock.


Often the information is incomplete and sometimes criminally so although it's hard to enforce laws in an area where there are few regulations and very little oversight.

They are essentially private contracts, so the two parties can put whatever in them they see fit. They are standardized to a large degree to allow a market to form around those contracts.


As we've seen trillions of dollars can seemingly quickly vanish. There's a lot of money to be made for a few people in the right spot at the expense of destabilizing the entire market.

This is just as true for the primary market.

You're certainly right that it can use more regulation and oversight, but saying that it's the derivatives market that is causing the primary market to lose forecastibility doesn't seem correct to me. It's easy to point to a little-understood secondary market as the culprit for problems that exist in the primary market, but we've had crises and stock market crashes long before we've had a large secondary market. Realistic forecasting can only exist in an inefficient market, if you're going to try to make the market inefficient you'll have to do a lot more than regulating derivatives.

caveman1917
2013-Mar-22, 08:07 AM
I was just reading something. Some outfit in Zurich has just published a paper that claims the 60 -70% of all moves in the commodities markets (which is what they studied, I think) are due to the system itself, and not any exogenous information.

Is the culprit "technical analysis" type self-fulfilling prophecy by the traders as Caveman above mentioned? Well, this seems to be the emergent behavior of HFT systems, which we've discussed in Black Monday on and off again. It was only about 30% "endongenous" before the rise of the machines. So, maybe humans themselves would tend to do it, but you need a computer to really go wild with it.

So heck, soon markets may be nothing but computer simulations....

Really, this HFT stuff needs to be stopped. If you're not familiar with, read about some of the analysese of the various "flash crashes" that have occured, where the computers went wild. There's a whole fascinating world of the emergent behavior of these systems that exist at the microsecond level. Anyway, there's several proposals (which aren't going anywhere because there's so much money involved with the HFTers) that would limit things to human reaction times.

But is HFT the cause of the loss of exogeneity? During the time of the rise of HFT we've also had the process where the markets became more accessible to small private investors (internet brokers come to mind). Where it used to be that the commodities market was primarily about firms buying and selling those commodities, it has, even without HFT, as a market itself become more open to small traders that trade on the price movements themselves, without interest in buying or selling those commodities as commodities.

HenrikOlsen
2013-Mar-22, 08:58 AM
Or is it simply that the market left the Greenspan mode and the belief in the effects of external controls has evaporated.

starcanuck64
2013-Mar-22, 07:13 PM
The models aren't that complex, you pay me say 1000$ for the right to buy 100 stocks of company X at price Y$ at the end of the week, can't get much clearer than that. The pricing models can get complex, but so are those in the primary market. Ask anyone for their model about what is the "right price" for a stock, it'll be just as complex as their model for the "right price" of a certain option on that stock.



They are essentially private contracts, so the two parties can put whatever in them they see fit. They are standardized to a large degree to allow a market to form around those contracts.



This is just as true for the primary market.

You're certainly right that it can use more regulation and oversight, but saying that it's the derivatives market that is causing the primary market to lose forecastibility doesn't seem correct to me. It's easy to point to a little-understood secondary market as the culprit for problems that exist in the primary market, but we've had crises and stock market crashes long before we've had a large secondary market. Realistic forecasting can only exist in an inefficient market, if you're going to try to make the market inefficient you'll have to do a lot more than regulating derivatives.

$680 trillion doesn't sound very secondary to me.

And if you look at the comments of Brooksley Born who was supposed to be in charge of enforcing some regulation in the trade of derivatives there was a large amount of fraud going on, there may still be if you look at what just happened with JP Morgan. It introduces a significant amount of chaos into an already chaotic system, while the potential for personal gain is huge, the overall result has been a destabilizing of the entire economic system of the world I think. If so much economic activity is happening in such an opaque and unregulated sector, how can you even state that there's meaningful economics going on at all. You have a system tumbling down into failure in my opinion.

http://www.pbs.org/wgbh/pages/frontline/warning/interviews/born.html


And also, we were seeing some very dangerous things happening in that market. There were some major fraud cases. There was use of over-the-counter derivatives to manipulate the price of commodities. And there were some spectacular failures by institutions that were speculating in the over-the-counter market with little or no restraint. For example, Orange County, Calif., was brought down, went into bankruptcy because of its speculation, gambling with public money in the over-the-counter derivatives market on interest rate swaps.

publius
2013-Mar-28, 01:18 AM
Re HFT (I debated whether to just post this in Black Monday, but since we got to talking about HFT here...). Nanex did an in depth investigation of the "Flash Crash" back around May 2010 (it's fascinating if you're interested, looking at things at the microsecond time scale), and they just released this:

http://www.nanex.net/aqck2/4150.html

IOW, computers going wild is a real problem. There is emergent behavior there that no one really understands or appreciates.

starcanuck64
2013-Mar-28, 08:30 PM
So the issue is the HFT firms provide liquidity slowly over the day as they buy stocks but they can demand liquidity(sell stock) very rapidly causing an imbalance in the system?