The Forecasting Game

Flattr this!

As a public figure, I’m quite accustomed to seeing my name and face in the media. But it was still quite a surprise to load The Sydney Morning Herald’s site on Monday last week and see my craggy visage staring back at me from the front page:

The reason was Peter Martin’s article A Keen eye on the global economy: how our forecasters fared”, which reported the results of the Melbourne newspaper The Age’s annual survey of economic forecasters.

Click here to read the rest of this post

About Steve Keen

I am Professor of Economics and Head of Economics, History and Politics at Kingston University London, and a long time critic of conventional economic thought. As well as attacking mainstream thought in Debunking Economics, I am also developing an alternative dynamic approach to economic modelling. The key issue I am tackling here is the prospect for a debt-deflation on the back of the enormous private debts accumulated globally, and our very low rate of inflation.
Bookmark the permalink.

24 Responses to The Forecasting Game

  1. scepticus says:

    Steve from the article:

    “Simulations of Lorenz’s model also strikingly illustrated how different a “linear, equilibrium” world is from a “nonlinear, far from equilibrium” one. If the model began in its equilibrium, it stayed there – as in this simulation in my modelling program Minsky…Compare this to what happens if one of the variables is a tiny, tiny distance from its equilibrium”

    Hi Steve,

    It seems that two things are confused here – chaos and ‘far from TE’ systems. With regard to the behaviour of Minsky you seem to describe a chaotic system, very sensitive to initial conditions. But a far from TE system (non equilibrium thermodynamic system, or NET) is created by an external gradient – an exogenous forcing – which for terrestrial weather is the solar gradient.

    So are you suggesting that the human economy is a NET, or just chaotic? The latter can utilise conserved variables with complex positive and negative feedback loops and memory can it not, while the former requires some kind of irreversibility in addition to an exogenous gradient.

    If the financial economy operates analogously or exactly as a NET, what is the exogenous gradient that drives flow in the financial economy and what is the location of and the dynamics of the irreversibility? Does Minsky include any processes which are non time-reversible?

    Cheers,

    Scepticus

  2. Steve Hummel says:

    May I suggest that two of the variables in Minsky be weighted appropriately in importance as Individual Purchasing Power (presently inherently scarce in comparison to prices by the reality of cost accounting convention) and Technological Innovation (inexorable in regard to efficiency of production).

    That way not only will forecasting be scientifically improved which is a good thing for profits and investors, but the system will become more humane because its intentions will have changed from the will to power of the various entities in it to the will to freedom for the individual.

    All realities are relevant and require attentiveness, but the higher instability/asymmetry needed to be aware of are the ones between Science and Wisdom, and Power (Control) and Grace (Individual Freedom).

  3. koonyeow says:

    Let’s turn economics from Ptolemaic to Copernican, Steve.

  4. Steve Hummel says:

    How about if we do both that and also turn it from Old to New Testament. I a non-parochial, modern fashion of course.

  5. koonyeow says:

    Pardon my ignorance, Steve H., but I can’t catch your analogy/metaphor.

  6. Steve Hummel says:

    Ptolemaic to Copernican is indicative of the paradigm change in regard to the nature of the cosmos itself.

    The non-parochial Old Testament (dead letter of the law/system) to New Testament (emphasis on the essential importance of spirit of the law/individual) concurrent paradigm change addresses that long standing asymmetry of power between system and individual.

  7. koonyeow says:

    I see. Thanks, Steve H.

    My analogy is actually more about the approach to economics, rather than a paradigm shift.

    By approach I mean: tinker, observe (for patterns), hypothesis, experiment, reject/refine the hypothesis, repeat.

  8. kalman says:

    There is a big differences between the USA and Australia, the USA had a problem with their Subprime lending, this was around 10% of loans, this played a part in expanding the bubble.
    Here in Australia we don’t have the 10% subprime problem, here we have around 80% of borrowers subprime and also our real-estate is around 250% more expensive than at the height of the USA Ponzi boom. We will have a devastating collapse of around 75% in nominal terms. Here on the Gold Coast QLD prices have caved in during the last two years, a minimum of 20% and some 60%.
    Please don’t think that people are flooding Australia buying our expensive produce, services or assets, that is only a real estate’s wet-dream, if you drive around you can see all the businesses going bust thanks to the overinflated AU$. Every country is doing its best to devalue its currency to help kick start the economy, it is amazing how stupid our government is to be the only place on earth to do the opposite, I can understand the average knucklehead citizen cheering this on, so they can spend more on eBay but they will lose their job because of Australian competiveness.
    This phenomenon manifested itself predominately as a result of our ” Lucky Country Status” meaning we had such a good life here, astronomical wages, 0 % unemployment for people who want to work, massive handouts, basically we are the Fat lazy sloth of the world. You see when you live in a Country where there is war or was in your lifetime, or a country where people starved to death like India, China, Pakistan most of the world, than you are a bit more humble and realize that you need to work for money not just expect it for nothing like during this Real estate Ponzi Scheme. Australia and Canada are the last ones still standing in the dark, we will realize during the next 20 years what the consequences are of this massive over spending and arrogance.
    If you can sell up now and walk away with a small loss just do it, it is better to be on Zero and free than to be bankrupt and a Zombie , you won’t even be able to rent after that!! If you can walk away with a profit than buy Precious Metals and rent for 5 years, wait until there is blood on the streets. Just like there is on every other country now.

  9. Steve Hummel says:

    Post to Mish Shedlock’s blog:

    Eliminate the “there” that is the source of the problem, i.e. the personal debt. That enables a reset. Then, in order to end the inevitable re-build up of debt institute a citizen’s dividend which eliminates the need to borrow to escape the austerity the system enforces. Remember, the middle class of America from 1946-1970….was an anomaly….not a normal feature of the system. Stop worshiping the system as it is, start correcting it.

    Economists must stop passing technological innovation and the reality of the effects on individual incomes that the rules of cost accounting enforce on the economy as mere externalities. They are operative factors on both the system and the individuals within the systems.

  10. Steve Hummel says:

    In response to Andrew Lainton’s very promising study here:

    http://andrewlainton.wordpress.com/2013/01/16/modelling-the-pin-factory-understanding-the-division-of-labour-using-a-stock-flow-model-part-1/#comment-8296

    This:

    Sounds like a potentially enlightening experiment. (It was actually what C. H. Douglas did almost 100 years ago.) Don’t forget to factor in the ever present commercially ENFORCED REALITY of the inherent scarcity of incomes in comparison to prices required by cost accounting’s rules. (profits of course are not wages, and only momentarily “solve” things for businesses while the ongoing effects are born by actual people) That factor and that fact always seems to get missed by economists (except by C. H. Douglas), and will make the REAL price inflationary nature of the system for BOTH individuals AND businesses apparent. Economics is complicated and has numerous possible problems and causes of problems, but money is basically accountancy and an unrecognized or unacknowledged flaw there, because it is a part of the very “woof and warp” of the system itself, is by definition a ROOT cause and will destabilize the economy and rob individuals of purchasing power.

  11. Steve Hummel says:

    Oh, and making a sweatshop, low tech example your model may lack a certain symmetry with present reality, but if you’re real conscientious about modeling the cost accounting effects as per above….the same century old conclusions ought to still be there.

  12. peterjb says:

    “Turning economics into a science” complete with $ sign

    http://www.youtube.com/watch?feature=player_detailpage&v=Gc9736gkFhw#t=126s

    This must be the most Vulgar and Crasse depiction of anything to do with economics that I have ever seen. And, a claim that a software modelling tool can convert a pseudo based religion into a Science.?

    And, downloading and running this “Minsky” I am advised, will clearly indicate a serious devolution in the program development and presentation, from the original software and code, which was called QED. In every aspect.

    I hope that those funding this work don’t expect that $ sign to become a profit reality.

    I repeat: Vulgar and Crasse.

    Ho hum

  13. Ted Stead says:

    Steve, good article, but just wondering where we are at with the Australian credit impulse/accelerator. Will you have an update of this any time soon?
    Cheers.

  14. koonyeow says:

    Title: Time Will Remove The Fragile and Maintain The Robust

    There was a time when classical thermodynamicists did not accept Boltzmann’s atomic interpretation of the 2nd Law. Of course, we now know that time has proven Boltzmann to be correct; just as time has proven that I am a skirt chaser.

    There was a time when people did not know why the tool is called Minsky. Of course, time will prove that Minsky was born out of Financial Instability Hypothesis and Monetary Circuit Theory.

    Steve will only be dis-credited if he turns neo-classical or Minsky is no longer open source (I will have to swallow one bottle of vodka in one minute if Steve turns neo-classical).

  15. Steve Hummel says:

    Why is it that the household economy is not completely analogous to the national economy? It’s because the individual cannot create and destroy money, but the FED/Private Banks/Government can do both of these….and then the FED/Private Banks/Government fail/refuse to keep that creation and destruction in stable balance by NOT enabling the individual to always have enough money to liquidate production as it comes to the market. It fails to do this by not financing the gap between total prices and incomes produced in each productive/financial cycle. Velocity does not sufficiently add individual income to rectify this ever present scarcity and hence the only way the system can continue is with the continual build up of debt. And a continual build up of debt as we see is very problematic.

    The FED/Private Banks/Government fail/refuse to do this because if it did allow such financing….the Private Banks, who not coincidentally are the most powerful of that triumvirate, would lose the greater part of the tremendous market that is consumer finance itself. And that wouldn’t be good for their bottom line. Also, they would actually lose control of the system…to the needs and desires of the monetarily and economically freed individual.

    Again, why do the mass of individuals not have the ability to liquidate production as it comes to the market? Because the effect of the rules of cost accounting is virtually ALWAYS a scarcity of total individual incomes in ratio to total prices WHICH IS CONTINUALLY ENFORCED WHENEVER, AND ACTUALLY BECAUSE, MONEY IS CHURNED THROUGH COMMERCE BY EITHER BANK LOAN OR GOVERNMENT SPENDING and so prohibits it from being possible EXCEPT AND ONLY MOMENTARILY under the most extreme and/or negative circumstances like huge government deficits during a depression or when the nation is engaged in total war. In addition to this, fewer and fewer individuals can actually find jobs because technological innovation is reducing the need for their efforts and inputs.

    And so we see that we have both a self interested minority ENFORCING an inherently unbalanced and increasingly unstable SYSTEMIC situation and an inexorable exacerbation of same due to the inexorable march of technical efficiency.

    And yet this could be rectified by simply changing the consumer financial paradigm from employment and loan ONLY to employment where available, UNIVERSAL DIVIDEND and loan if desired and creditable. This would be a free and free flowing economy as well as the individual set free. Furthermore, the individual would be much more empowered to determine who remains in business through the exercise of his/her sufficient purchasing power money-vote rather than the self interested minority of financial and corporate entities like we see presently.

  16. Lyonwiss says:

    Peterjb January 17, 2013 at 11:35 am

    Economics is not, and has never been, a science. Economists do not understand what science is. They merely imitate physics with mathematical models. But without anything remotely like a law, having the certainty and precision of those in physics, economics cannot possibly make forecasts, because the models have to be based on real laws to have any deductive validity. I’m writing an academic paper on this subject.

  17. Steve Hummel says:

    Money is basically accountancy. If the rules and effects of accounting, specifically the rules of its subset cost accounting, are not recognized and compensated for then confusion, authoritarianism and half @ssed theory arises. Most “emergent” qualities in economics arise as a result of failing to model these ever present monetary/accounting effects. Economics itself has many things which do have to be modeled in the attempt to forecast and invest, but if the above most basic facts are not modeled/accounted for…..then there is instability and consequently forecasting is difficult.

    Last but not least the humanity of the system must be stressed more so than even the stability and work-ability of same. This has always been a problem which has been neglected and even denied by the more culturally hidebound. The system can never be completely free without the individual being so as well. The best way to insure that our economic/financial/monetary systems are humane is to base them on the deepest condensations of human wisdom instead of lesser abstractions like profit or employment. Profit and employment are fine as secondary or tertiary considerations, but again, they pale in comparison to the deepest of wisdom’s condensations which are the ideas, values, purposes and most importantly the experiences of Faith as in Confidence, Hope, Love and Grace. All of the world’s wisdom traditions will be found to have these or their cultural equivalents.

    And the process of deriving policies from these is no more difficult than deriving them from profit or employment as primary ideas. The only thing that is really necessary is to be thorough and consistent in doing so. Man’s species designation is homo sapiens, wise and discerning man. It’s time we crafted systems aligned with and based on that wisdom. The better to create more of the fruits of such both individually and systemically.

  18. Bhaskara II says:

    Kalman,

    I found your commentary on the state of the economy and your opinion on the future interesting and novel from media reports. If I remember right, you might have seen economic problems first hand unfold in various times and places in the world.

    Certainly one could be concerned about high purchase priced highly leveraged real estate, for sure. We have recently seen the problems become obvious in other places.

    On purchasing precious metals now:

    I included some gold charts and links. Some charts are logarithmic. In logarithmic charts exponential growth follows strait line with the slope indicating the exponential growth rate. In the 1970s decade gold went up to about $850 or so and near 1980 it plunged and went down with some wiggles for two decades. (Past performance doesn’t guarantee future performance. ) Yet in the 1980s and 1990s we certainly had inflation in the positive range, prices went up significantly in the two decades. So, it is possible for gold to go down in addition to up. Do you have a reason that it might be safe to put value in now with out it going down significantly? If one had bought near the top one could quickly have lost a significant value.

    http://en.wikipedia.org/wiki/Gold_as_an_investment Two gold charts

    Also, I think in 2008 lots of things went down temporally including precious metals. Platinum dropped spectacularly. I think platinum went down even more because car manufacturing stalled and it is used in catalytic converters for cars. But I’m not sure why the drop. Could the catalytic converter folks have had to sell their stocks of platinum? Not that things would necessarily play out the same way.

    In our case now do you think precious metals is definitely not going to go down a lot but keep rising? Rather than drop some time? Do you think it is a safe speculation from here? I’m eliciting your opinion on what, when, where, how, and why? Have you seen some thing similar and does it apply to the US, Australia, other places today.

    If gold does go or currency down spectacularly, or other problems do you think there would be any thing to notice ahead of time? Have you seen such indicators in some other places?

    If your short of time you could name the event and or time and place and I could go read up.

  19. Bhaskara II says:

    Kalman,

    I found your commentary on the state of the economy and your opinion on the future interesting and additionally novel from media reports. If I remember right, you might have seen economic problems first hand unfold in various times and places in the world.

    Certainly one could be concerned about high priced highly leveraged real estate, for sure. We have recently seen the problems become obvious in other places.

    On purchasing precious metals now:

    I included some gold charts and links. Some charts are logarithmic. In logarithmic charts exponential growth is on a strait line. The slope gives the exponential growth rate. In the 1970s decade gold went up to about $850 or so and near 1980 it plunged and then went down with some wiggles for two decades. (Past performance doesn’t guarantee future performance. ) So, it is possible for gold to go down in addition to up. Do you have a reason that it might be safe to put value in now with out it going down significantly? If one had bought near the top one could quickly have lost a significant value.

    http://en.wikipedia.org/wiki/Gold_as_an_investment Two gold charts
    http://www.newworldeconomics.com/archives/2011/060511.html A long run commodities chart.

    Also, I think in 2008 lots of things went down temporally including precious metals. Platinum dropped spectacularly. I think platinum went down even more because western car manufacturing stalled and it is used in catalytic converters, but I’m not sure. Could the catalytic converter folks have had to sell their stocks of platinum or they sold high earlier? Not that things would necessarily play out the same way.

    In our case now do you think precious metals are definitely not going to go down a lot but keep rising? Rather than drop some time? Do you think it is a safe speculation from here? I’m eliciting your opinion on what, when, where, how, and why? Have you seen some thing similar and does it apply to the US, Australia, other places today.

    If gold does go up or currency down spectacularly, or other problems do you think there would be any thing to notice ahead of time? Have you seen such indicators in some other places?

    If your short on time you could name the event time and place and I can go read up.

  20. Bhaskara II says:

    Kalman,

    I found your commentary on the state of the economy and your opinion on the future interesting and novel from media reports. If I remember right, you might have seen economic problems first hand unfold in various times and places in the world.

    Certainly one could be concerned about high priced highly leveraged real estate, for sure. We have recently seen the problems become obvious in other places.

    On purchasing precious metals now:

    I included some gold charts and links. Some charts are logarithmic. In logarithmic charts exponential growth is on a strait line. In the 1970s decade gold went up to about $850 or so and near 1980 it plunged and went down with some wiggles for two decades. (Past performance doesn’t guarantee future performance. ) So, it is possible for gold to go down in addition to up. Do you have a reason that it might be safe to put value in now with out it going down significantly? If one had bought near the top one could quickly have lost a significant value.

    http://en.wikipedia.org/wiki/Gold_as_an_investment Two gold charts
    http://www.newworldeconomics.com/archives/2011/060511.html A long run commodities chart.
    {Trying to put up other graphs}

    Also, I think in 2008 lots of things went down temporally including precious metals. Platinum dropped spectacularly. I think platinum went down even more because western car manufacturing stalled and it is used in catalytic converters, but I’m not sure. Could catalytic converter folks have to sell their stocks of platinum or sold high earlier? Not that things would necessarily play out the same way.

    In our case now do you think precious metal are definitely not going to go down a lot but keep rising? Rather than drop some time? Do you think it is a safe speculation from here? I’m eliciting your opinion on what, when, where, how, and why? Have you seen some thing similar and does it apply to the US, Australia, other places today.

    If gold does go or currency down spectacularly, or other problems do you think there would be any thing to notice ahead of time? Have you seen such indicators in some other places?

    If your short on time you could name the event time and place and I can go read up.

  21. Bhaskara II says:

    Very sorry about repeats, I was getting errors and no indication of a post when posting.

  22. Bhaskara II says:

    Gold Chart(s):

  23. Bhaskara II says:

    Kalman,

    Haven’t been able to upload charts. No matter link could suffice.

  24. Bhaskara II says:

    Kalman,

    Here are plots links I found to replace the plots I could not upload for you.

    Subject: Historically gold can go down in addition to up:

    Gold log-plots: Rise before 1980, peak near 1980, and a fall after 1980. An example of a time when gold didn’t go up forever (in US$) and actually went down later that we might have a memory of. For log plots need to read the scales to see whats going on.

    http://gold.approximity.com/since1968/Gold_USD_LOG.html
    Notice the distance from 1980 top to bottom near 1982 is the same as the distance between the 500 and 200 mark on the vertical scale. So (trough price/ peak price)=200/500=2/5=40% of peak. Notice the scales double for the same vertical length, look at the distances between 50,100,200 and 500,1000, 2000 all the same distance for 2 times. Or same distances n times distance. The bottom of the graph is near 25 a half of 50.

    http://goldsilverworlds.com/wp-content/uploads/2012/07/gold_price_chart_200_years_1800-2012.gif Shows rise-peak-falls in 1815, 1863, 1940s and 1980. And rises near 1834, 1933, the decade of 1970s,and the decade of 2000s. After 2013 what could gold do? I can’t tell the future. Exponential rises in 70s and 2000s are almost constant slope lines. Slope indicates constant % growth rate per year, equivalently constant exponential growth rate. Explanation of log plots are in high school higher algebra textbook or higher math books in the section on logarithms and exponents.

    http://goldsilverworlds.com/wp-content/uploads/2012/07/gold_price_trendline_2000-2012.gif
    Plots price 2000-end of 2012. The price is almost doubled every 4 1/3 years. Or, had a near constant growth rate of 17.4% per year growth compounded along the strait line. For a total gain of times 1700/250=6.8 times the near year 1999-2000 low. Future may or may not reflect the past. You can see gold price has moved approximately horizontally after the 2011 peak.
    ————————————————-
    Why use Log-scale graphs in addition to linear scale graphs?

    1. Shows constant % growth rate as a sloped strait line. Or, constant exponential growth. A higher growth rate is steeper upward sloping line. Faster decay is a steeper downward sloping line
    2. The same vertical distance on a plot is a constant multiplier when going up or a constant divisor when going down. So, if it goes down by 1/2, to go up 2 times would be the same distance upward. That would be the same for any divisor/multiplier such as 1/10th and 10 times.
    3. Good for long term time series with exponential growths. It will stretch out the the lower end and compress the upper end. Exercise compare Dow-Jones index in log scale to regular linear scale. You will see no detail in the distant past in the regular linear scale plot. This problem mainly has to do with constant inflation rates actually being an exponential effect.
    4. Warnings: Usually an upward slope does not go on forever for real things measured in real units.
    5. Strangeness: There is no, 0, zero. Because there is no log(0). As you go down you just divide by a number for each distance which will get close to zero but never will.
    6. More info on log plots can be found in an algebra text book or higher math that has a section on logarithms.
    7. Log plots many times can be really cool to use especially looking at the past. But, predictions from them using exponential rates are highly suspect. Nothing real goes up exponentially forever. On could be tempted to draw a strait line into the future that will could never happen. That is because on the log plot a strait line looks linear but it actually represents exponential growth. This could be demonstrated by looking at the debt to GDP ratio on a log plot before, during , and after the 1930s, and 2008.

Leave a Reply