--
You received this message because you are subscribed to the Google Groups "LibertarianSA" group.
To post to this group, send email to li...@googlegroups.com.
To unsubscribe from this group, send email to libsa+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/libsa?hl=en.
You're being unfair to economists, Trevor. Surveys find that there's a high level of consensus amongst economists on most issues. The impression that there are muddled economists arises mainly because media present "both sides", which leaves the impression that nutters like Krugman and Stiglitz are representative.
As for clarity, most economics I've read are as clear as intellectuals get, eg Freakonomics and dozens of other books/articles.
Economics should be distinguished from econometrics and mathematical economics, which, in turn, differ fundamentally from each other. The two of them are also typically quite clear, provided one is into quantification or mathematics.
Although, having written that, I should add that what econometricians write tends to be clear, provided you make allowances for their curious methodology and assumptions. Outside their logic bubble, much of it garbage.
There are many examples. Take the way they calculate GDP. That's their way of saying how much wealthier (or poorer) we are. Mainstream econometrics (as opposed to what comes from our guys) is what gets Krugman spouting such absurdities "seriously" (his word) as the need to spend as if as if we’re expecting an alien invasion -- that's his crypto-intellectual version of digging and filling trenches to create wealth. It also gets muddle-headed commentators saying superstorm Sandy, wars and other horrid stuff that goes bump in the night, "stimulate" growth.
No, that’s not a typo; it’s really what they say. Such sophistry entails variations of the old broken window fallacy, which vary in form, not substance. Likewise beneficiation babble.
It goes like this: when you build a house it's added to the GDP. That it's you, not the rest of us, who has the house doesn’t occur to econometricians, but that's another matter. Collectivism and macro-economic aggregates presume what's yours is ours. I like that. You being richer somehow makes us richer, and the main responsibility of econometricians is to keep telling us that.
Anyhow, when your house gets blown away by three little piggies or a tornado, you'd expect econometricians to deduct it from the GDP, but they don't. In fact, they add it in again when rebuilt. They create the illusion that the country is a house richer whereas it's a house poorer. You have to give them credit for looking on the bright side when the pawpaw hits the air conditioner.
There's also the seen and unseen; we see the rebuilt house, but not what else there would have been had resources not been diverted. In case you're thinking econometricians are remiss, take back your naughty thoughts. It's simply and obviously completely impossible to know or even speculate how else the resources would have been deployed. Would there have been more tooth picks, jumbo jets, wrestling matches, game drives, promiscuity, moon landings, radio channels, tweets, widgets, woblets, wangles, globifers or waggloplets? Or lunches. Since no one has the slightest idea, it's best to ignore alternatives and look only at what we have, which is a new house. To point out that there's no such thing as a free lunch is mean and nasty and spoils the fun of the numbers game.
We don't really have a new house, of course -- we have the same number we had before -- but, with all this building going on, it's unreasonable to ask econometricians to say nothing happened, nor to expect them to sow doom and gloom by saying we're a house poorer when there's a new house for all to see. No one would believe them anyway, and they'd lose credibility. Nor can we ask them to muse about what might have been, nor (as micro-economists do) to say it's your house. It's more congenial, after all, to say we're all in this together, and to prove it with impressive numbers and acronyms.
Another example of econobabble is international trade. Everyone knows that if you get more than you give, you have a surplus. Yet econometricians call precisely the same thing when it happens internationally a "deficit". Getting more than you give for them is being worse-off. Yes, I know that's weird, so I’ll explain how they perform the smoke-and-mirrors illusion in a moment.
Meanwhile, ordinary folk also know -- or would were they to think about it -- that all voluntary exchanges add value; they leave all parties wealthier. Otherwise they wouldn't trade. That’s obvious. Well, not really. Accountants aren't quite so cheerful, so they don't see a perpetual positive sum game. For them everyone's goofy because they go to all the effort (transaction cost) of trading at the end of which all their books balance to the nearest cent, which means all transactions are always in precise equilibrium and therefore a complete waste of time.
If you think that's weird, you ain't seen nothing yet. Enter econometricians. They're really amazing; they add all the participants’ pluses, and all the accountants' zeros, and, as if by magic, they declare a deficit.
How do they achieve this astounding feat? Unlike everyone who trades, and all accountants, econometricians attach zero value to what's traded. That's right, for them, what people do (internationally at least) is pass money around in a kind of global parlour game during which econometricians take occasional snapshots. At the end of fiscal years, or quarters, or whatever, if everyone collectively in country A paid more to people in country B than they got from them, it's called a "deficit". Admit it, you can't argue with that. Country B gained money at the expense of country A.
Since no one questions the assumption that international trade is a money-circulating parlour game, no one blows the whistle of the surplus-deficit illusion. Anyway, econometricians are really clever, so no one dares argue with them even if the nation’s experience isn't theirs. We assume clever people would never do something so manifestly nuts as to ignore why people pass money around.
In their defence, econometricians have to do what they do, you see, because if they don't, like accountants, they'd keep reporting zero balances, and they’d be boring and/or unemployed. Instead, they've create elaborate schools of thought, and win Nobel prizes, and churn out complex theories about how countries can get more money than they give, whereas everyone else wants to know how to get more stuff than they give; how to get more for their money rather than how to get more money per se.
So, Trevor, be fair, and give credit where credit is due. On reflection, don’t give credit; you might create a debt crisis. I'll explain that too if anyone (a) reads this far and (b) wants to know how econometricians turn lots of individual debts into "national" debt. There is, of course, no such thing, but unless we assume there is, we'd have no need for pseudo-intellectual mumbo-jumbo about national debt? Just forget it all, Trevor, and enjoy the clear congenial economist who pleased you ... and me ... thereby increasing the GNH. That's econometrese for gross national happiness.
ciao
Leon
Enjoyed your piece Leon J
And by the way I often wonder if Americans ever read the story of the Three Little Pigs.
Can anyone tell me why they build their houses planks? Surely if your house is in an area that might be hit by a tornado or such, you would want to build it sturdily with bricks. There must be some technical reason why wood is preferred. If it was only for the initial economic reasons then surely the insurance premiums would be higher for wooden houses than they would be for brick house. Does anyone have any information on this.
“. . . Anyhow, when your house gets blown away by three little piggies or a tornado, you'd expect . . .”
--
--
OK, but you would think that by now they would figure out that cheap & quick just doesn’t stand up to stress.
Where are all the American great inventors to discover new durable materials with which to build houses?
J
I think the opposite is true, that our regulations are too rigid, and we should build houses out of materials other than brick and concrete.
The SA building regulations interest with the free market where people could have more free choice about which materials they may build their houses.
This would lessen the (artificial) housing shortage.
Those asbestos houses were dirt cheap, sturdy, quick to construct... but some bureaucrat decided it was dangerous and outlawed using asbestos as building material. It must have been in the late 80's, because I grew up in a house made of asbestos panels in the 1970's and we had classes at school in the 1980's made from them. Those houses and classrooms were warm in winter, cool in summer, are still standing and still look like they did when they were built. I think a huge injustice was inflicted on the country "for our own good". So now you can't breathe in asbestos fibres, but schooling is under a tree and the roof over your bed a traffic sign. I don't think we are better off for it, but the argument against asbestos can be seen, the houses & classrooms not built from it cannot be seen. What is seen usually wins over what is not seen... like the Gautrain won over all the other things that money could have bought but didn't.
S.
Correction: "harmless" should read "harmful" and "I" should read "It".
S.
-----Original Message-----
From: Stephen vJ (Gmail) [mailto:sjaar...@gmail.com]
Sent: 12 November 2012 09:42
To: 'li...@googlegroups.com'
Subject: RE: [Libsa] I'll Huff & I'll Puff!
Well, I'm not saying it is not dangerous. I might yet die of some kind of lunch disease. What I'm saying is that people are not given the free choice between using a dangerous building material and all the costs associated with it vs. the alternative, which is often being homeless / schoolless. Knowing how bad it is, they might choose to cover themselves with road signs and be schooled under a tree. They might not. Imposing the bureaucrat's subjective value system on people means that all those who would have taken their chances with a harmless substance are deprived of what they see as value exceeding those risks. It causes decisions to be made based on the subjective value judgements of a few rather than the subjective value judgements of everyone.
S.
-----Original Message-----
From: li...@googlegroups.com [mailto:li...@googlegroups.com] On Behalf Of David Joffe
Sent: 12 November 2012 05:04
Wow, three spelling mistakes in one short post. At least my logic is solid. ;-)
S.
Hi Trevor
Economics is a social science. Theories are tested by empirical evidence which is itself subject to statistical variations. The fact that economists try to explain the real world of human behaviour, does not make economics any less scientific. Psychology is another social science where human behaviour scientists try to explain human behaviour, and subject their theories as far as possible to double blind statistically valid tests and samples of empirical evidence.
There is such a thing as water, and scientists can agree that is is composed of oxygen and hydrogen.
However there is no such thing as a recession. It is merely a construct that economists have defined to try and explain real world behaviour. As such there is much debate as it is impossible to put "recession" in a laboratory. Besides many made millions in the great depression.
Last point, physicists don't agree at all, as you have stated. There is vigorous debate about string theory, dark matter etc etc
Economists agree on most of economics, psychologists agree on most of psychology, and physicists agree on most things physical. But where any of the disciplines try to explain behaviour, be it of people or the cosmos, where that behaviour lies on the edge of current human knowledge, then there is still debate.
It is misleading therefore to single out economics as a science where the experts disagree.
--
--
--
--
Here's the cause of the trouble. The typical economist will spend one long evening pulling an all-nighter to pass Economics 101. In that one long evening, one question / fact which is covered, amongst many others, is that GDP as an economic measurement has some shortcomings. One of those shortcomings, depending on which text-book is used, might be that it does not take existing wealth into consideration i.e. the house that you paid off 10 years ago but still derive economic benefit from by living in every day is now shown in GDP. Those textbooks who list that shortcoming will typically note that existing capital is not shown because it is not traded i.e. it has no price (but has value) and therefore cannot be measured. Now, that shortcoming of GDP is a one sub-part of a small part of one long evening. That same economist will now spend the rest of his academic career studying ways to increase economic growth as measured by GDP, determining the impact of monetary policy on GDP, figuring out how changes in GDP affects longevity, what role education plays on GDP and so on. So when Sandy destroys the house that was paid off 5 years ago and GDP remains flat or even increases, do you blame the economist for focusing on the measure his whole academic career has been centred around, forgetting that one factor in that one definition of that one evening back in the early days just before his Economics 101 exam ? Please. Have some compassion.
S.
From: li...@googlegroups.com [mailto:li...@googlegroups.com] On Behalf Of Gareth Brickman
Sent: 13 November 2012 09:06
To: li...@googlegroups.com
Subject: Re: [Libsa] Fwd: An economist who speaks sense - see attached article
@Garth
--
--
I think economists are very good at predicting the future. It is your expectations of their predictions which is wrong. Let me demonstrate.
If you want an economist to tell you when a certain commodity price has hit a peak or what the price of butter will be relative to guns, that is just like asking a meteorologist to tell you what the temperature will be on Christmas day. You are expecting omniscience from a mortal in a complex system. It is not that their disciplines are lacking something or that they are stupid & ignorant, but that you expect them to accomplish the impossible.
A meteorologist can tell you other things i.e. that it will rain less in 2014 than in 2018 because of the El Nino effect, but they cannot tell you if it will rain on Christmas day. It just happens that the latter is personally more important to the party planner deciding whether to pack shorts and a towel.
Similarly, an economist can tell you exactly what will result from certain actions like price fixing = shortages. In Zimbabwe a few years back, they fixed the price of bread on Friday and by Tuesday the shops were empty - 100% hit rate on that prediction. Economists can predict a surprising number of results coming from particular events... what will happen to price if a new import duty is imposed or that the black market in horns will increase if you outlaw hunting, etc.
What people can predict is not a reflection of the accuracy and sophistication of their profession as much as it is the layperson's expectations of the extent to which those predictions can be applied to the daily problems they face.
S.
I disagree... and this is rare for us, so it must be a very special case. ;-)
An economist who says that the market rose today because people were feeling jittery about Greece must surely know that some individual sold because his wife ran off with the pool boy and he expects some unforeseen expenses to pop up shortly at a nearby gun shop. Greece is not and can never be the only reason or the reason for everyone... but it can be the major reason for most traders.
Every economist must also know when saying such things that for every seller there must necessarily also be a buyer and thus half the market feels the current price is worth paying and shouldn't go down much more i.e. feeling more positive about Greece than the other half. Every reason is therefore nil on net. When the whole market moves, he is correct in asserting that the one side's reason dominated slightly, knowing that the other side still exists.
The economist who ignores the major reason because there are also a myriad of lesser reasons by declaring his inability to know all of them will never be asked anything by the media again and will in fact be wrong. There is a major reason and he does know it. It is Greek jitters. Not saying so makes a fool of him in front of cameras and colleagues. Admittedly, he should end his sentence with "T&C's apply, not for sale to non-economists under the age of 86", but that is implied in the same way as "80% chance of thunder showers tomorrow" comes with T&C's.
"After all, how many main stream economists are vocal opponents of minimum wages, petrol pricing, bread price fixing, affirmative action, etc, etc?"
Maybe I only interact with some really special people, but I don't know a single economist who supports any of these on economic grounds. In fact I don't know any economists who support any of those on any grounds, except for one economist who supports affirmative action and even he admits that it is for socio-political reasons rather than economic ones and is quick to point out the economic down-side of it.
Oh, that makes sense. I once asked a professor in economics why the curriculum was so thin on Austrian economics and he said something like... do you really propose that we advocate to students that the government, who is by far the biggest financial backing of this university, stop its funding ?
I guess that is like volunteering your boss for a pay-cut and could get you transferred to the psychology department.
That said, I fail to see how this reflects on economists any differently than on the rest of us who use public road, pay our taxes, stop at traffic lights or RICA our cell phones. We all act in our own best interest according to the incentives found in our environment. Economists are no different in that respect.
S.
--
I'm game... except you need to clearly distinguish between economists on the one side and bankers, traders, market analysts, etc. on the other side. I have rarely seen economists make comments on daily stock market movements. Those are usually provided by traders or market analysts.
There's a need to work through the weird mythology that obfuscates international trade/finance. I'm spending time getting to the bottom of it with a view to (a) understanding it and then (b) communicating it. Needless to say, there is and should be a heavy onus on anyone who challenges the established view.I put some of my critique to a workshop with economists recently, including one who mastered in international trade and finance. They were perplexed, and none challenged me. The trade economist said he couldn't fault what I said, but he'd have to digest it later because it contradicted too many basic and unquestioned assumptions, formuli and articles of faith, eg that there are such things as national debt and trade imbalances, that getting more than you give should be called a deficit, and so on.You'll be pleased to know that you magically become a co-beneficiary of someone else's exports, but also that you're jointly liable for the imports and foreign debts of everyone who incurred liability without your knowledge or consent. How and when it happened is a mystery, but be assured that it is so ..... according to the national accounts. You might not have known it, but "we" have a balance of payments constraint.thus constrainedciao
@GarthHave you seen Krugman's story about the babysitting co-op? He describes it as having "changed his life" and that he thinks about "often".
Read his account of it, because it seamlessly illustrates the breadth of his ignorance when it comes to economic fundamentals. Now, I'm not saying Krugman is wholly ignorant or doesn't have economic insights to add (his work in the 90's on trade is very good), but I think it's drastically under-estimated just how ill-informed and ignorant even esteemed economists are of the basics.Austrian economist Jorg Guido-Hulsmann pointed out at a conference recently that basic economics is often taught with the later learning more intermediate stuff in mind, which is predominantly all about modelling. He says that in this framework very often students will get through their basic economics courses without having much understanding at all of the fundamentals, and are not able to really grapple with the different schools and different perspectives that are all about fundamental divergences of theory. I think the same goes for the guys at the top, so to speak, and which is why I very often don't see so much of a debate of principle and theory between differing economists as semantic and perspective confusion. The debate surrounding the eventual effects of Hurricane Sandy is a great example of this.
On Monday, November 12, 2012, Garth Zietsman wrote:
There is the IGM economic experts panel here - which is intended to give an idea of the economic consensus. The panel samples elite economists from a diverse range of ideologies. It seems there is a great deal of consensus on a wide range of supposedly controversial economic questions which confirms what a number of surveys have established. For what it's worth I have agreed with the consensus position on every question they have answered so far and I am pretty sure Krugman would have as well.
No doubt you can exclude me - I'm not an economist - but you are crazy if you want to exclude Krugman from the economic mainstream. The libertarian economists at George Mason University i.e. Tyler Cowan, Alex Taberock, Bryan Caplan etc all take him very seriously (while often disagreeing). In fact Caplan recently posted that he cringes every time he witnesses some libertarian claim that Krugman doesn't understand economics. The truth is the Keynesian models he espouses are definitely PART of mainstream economics (just as Austrian economic theories have been absorbed into mainstream economics), whether the fellows at Mises.org get that or not. Keynes had some real insights and wasn't by any means a nut. Hayek and Friedman at least could recognize that.I figure if we are going to seriously advance liberty and freedom it would be helpful if we didn't align ourselves with, or limit ourselves to, fringe economics - while thinking of it as mainstream. Those multi-prize winning and widely respected liberal economists that are being so cavalierly dismissed on this thread, are clearly not nutters. We do ourselves, and the libertarian cause, a disservice if we fail to understand what those guys (and much of mainstream economics) are really saying.
Garth
On Mon, Nov 12, 2012 at 9:48 AM, Trevor Watkins <bas...@gmail.com> wrote:
Hi LeonI fear that you prove my point for me - there is less consensus amongst "economists" (if you include Krugman, Zietsman, etc in this term) than there is amongst, for example, physicists or "Hard science" experts. Of course, those who agree with each other do indeed agree with each other.--
Trevor Watkins
On 9 November 2012 21:52, Leon Louw (gmail) <leon...@gmail.com> wrote:
You're being unfair to economists, Trevor. Surveys find that there's a high level of consensus amongst economists on most issues. The impression that there are muddled economists arises mainly because media present "both sides", which leaves the impression that nutters like Krugman and Stiglitz are representative.
--
You received this message because you are subscribed to the Google Groups "LibertarianSA" group.
To post to this group, send email to li...@googlegroups.com.
To unsubscribe from this group, send email to libsa+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/libsa?hl=en.
--
You received this message because you are subscribed to the Google Groups "LibertarianSA" group.
To post to this group, send email to li...@googlegroups.com.
To unsubscribe from this group, send email to libsa+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/libsa?hl=en.
When economists do that, I am certainly right next to you with a torch and a pitch fork (or a gun and a pack of sandwiches, as I believe the new version of that phrase goes). I am just a bit more reluctant to label any old hack an economist and find few real economists actually warrant torching.
S.
Hi LeonI fear that you prove my point for me - there is less consensus amongst "economists" (if you include Krugman, Zietsman, etc in this term) than there is amongst, for example, physicists or "Hard science" experts. Of course, those who agree with each other do indeed agree with each other.--
Trevor Watkins
On 9 November 2012 21:52, Leon Louw (gmail) <leon...@gmail.com> wrote:You're being unfair to economists, Trevor. Surveys find that there's a high level of consensus amongst economists on most issues. The impression that there are muddled economists arises mainly because media present "both sides", which leaves the impression that nutters like Krugman and Stiglitz are representative.
You received this message because you are subscribed to the Google Groups "LibertarianSA" group.
To post to this group, send email to li...@googlegroups.com.
To unsubscribe from this group, send email to libsa+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/libsa?hl=en.
Correction: "greed and fear" should have read "supply and demand".
You see, I'm so tired, I'm resorting to correcting other people's mistakes instead of my own. How silly of me.
Time for bed. Night all.
Schlichter's issue is not with mere fiat money- you'll notice he's fine with free banking including fractional reserve? No, the fundamental problem is when fiat money is imposed by govt as the sole monetary unit, usually with either the govt or a subservient organisation ordained as the monopoly producer.
As I said earlier in this thread, I see these sorts of discussions all the time and they never gain much traction because of semantic and perspective confusion. For example, when you say "disruptions due to mild inflation will be mild at worst and strictly temporary". What do you mean by inflation? Do you mean an expansion of the money supply or a general increase in prices in a part of the economy? What kind of money supply do you mean; what sort of banking and monetary system is this happening in? What do you mean by inflation being "mild"? Do you mean a limited expansion of the money supply? Relative to what? How "temporary" is "temporary"?
It's important to get the detail and the context right, rather than speaking too broadly and assuming the other person knows what we mean, otherwise we'll end up chasing our own tails around.
I say this because in your "specific source of disagreement" paragraph you've ended up conflating separate issues. Schlichter was clear in the section you question regarding inflation that he was framing his explanation in the "modern system"; ergo centrally-banked fractional fiat. To wit (emphasis mine): "Money injections have no lasting benefit for the economy; they always distort relative prices and reallocate resources and thus change the distribution of income and wealth. Money injections always create winners and losers. In our modern system they cause near-term booms followed by busts later, although complete busts are usually avoided for some time through accelerated money creation (monetary 'stimulus') from the central bank during recessions. As a result, the economy accumulates substantial misallocations of capital and other distortions over time that require ever more money-printing to stop from unraveling."
At the end of your paragraph, you write: "Elsewhere he states that a constant supply of money is fine, even in a growing economy, as this will result in a mild deflation. Somehow while the differential price changes via a mild inflation cause serious disruptions, the same thing via deflation does not???" The part where he talks about secular deflation is in response to the criticism that elastic forms of money are a necessity for production. He's describing how, by the market process, the exchange value of a fixed supply or unchanging quantity of money will simply rise relative to the increase of production, therefore constant production of money is not required if prices are allowed to adjust. Under the market process, this is not a problem- see, he's talking about a different environment to the centrally-banked fractional reserve fiat system above, and he's talking fundamentals of money, not giving the effects of monetary deflation preference over the inflationary boom-bust cycle.
"The story is a perfect illustration that lack of demand CAN cause a recession, even where there are no structural problems or misallocation of resources and everyone is reliable and wants to work."
Unfortunately, the babysitting co-op did have: structural problems; resources couldn't be co-ordinated rationally or efficiently because there was no price mechanism; and whilst the participants were willing to work, the system failed to account for their subjective valuations and preferences.
See, the babysitting co-op analogy only stands up if you take its constraints as well as "technical problems", as Krugman calls them, as given. Now, this is very similar to how many can come to misunderstand Keynes' prescriptions for remedying the specific conditions of America's depression in the 30's, and I have actually defended Keynes in light of this, much as his friend Hayek did (Hayek said the problem with Keynes' General Theory was that he called it a general theory, rather than making clear it was for specific circumstances in a specific time).
Nevertheless, here's my critique, and I don't think it requires a particularly Austrian approach:
Krugman, I would say disingenuously for his readers, seeks to exaggerate this analogy to whole economies (totally ignoring the vastly different fundamentals between the nature of the scrips and the nature of money), cynically waves away the fundamental alternatives (which he is biased against anyways), and cheekily hints at undesirable consequences of the prescription but does not elaborate ("Eventually, of course, the co-op issued too much scrip, leading to different problems.")
The fundamental problem with the scrips was that they assumed every half-hour of babysitting had the exact same value no matter when it was performed. This price control meant that there would naturally be shortages when demand was too high and surpluses when it was too low. The very nature of the scrips would also inhibit supply because it's likely demand would converge at similar times (when the participants would likely all wish to use them and have the opportunity to, like weekends, holidays etc.). See, the problem is the scrips are a lousy medium of exchange. Their exchange-value can't adjust to reflect the varying preferences of those supplying and demanding them; they don't have a price to reflect the prevailing market conditions at a given time. They're a beautiful example to illustrate the important characteristics of a medium of exchange, because as eloquent a solution as they seem to be, there're really just a very constrained form of barter.
The "technical problem" was that the operators of the co-op collected dues in scrip, which ultimately led to the lowered circulation of the coupons which would naturally exacerbate inherently screwed-up supply and demand dynamics.
The organisers of the co-op created their own problems, and even Krugman vaguely admits that their prescription to increase the supply of scrips eventually had more negative outcomes. So, rather than act as an analogy to shill for monetary easing, I think it does a pretty good job highlighting the deficiencies of central planning and reactionary interventionism to such deficiencies.
Oh, and lastly, I don't get the controversy over Say's Law. To me, it simply means that one’s ability to demand goods and services from others derives from the income produced by one’s own acts of production. Is that agreeable or do you have another take?
Have a good night,
Gareth.
In that case, Popper did the world a great injustice. That notion probably set the study of human behaviour (and attempts to predict it) back significantly. Luckily, neural science has advanced since Popper's time so that specialists in the field can now predict by means of brain scanners exactly what motions your hands will perform before you become conscious of them (1). It can even predict the frequency with which you will fail to add the "s" suffix to an accusative verb in sentences like "The man walks" just by looking at a particular bit of genetic material from elsewhere in your body (2). It can work out that you are about to have an eureka moment up to 150ms before you actually have it (3). Human behaviour most certainly can be predicted, even moral behaviour which happens on genetic autopilot most of the time rather than through rational thinking as previously thunk (4). Having read about all of this, I know just what chemical reaction the word "thunk" just generated in your mind. Shame. Sorry. The most recent, hard, scientific evidence is solidly against Popper. At least Mises had the insight to say that, as he was writing it in 1941, that human action could not be predicted yet and that mathematical modelling of behaviour could not be done given the tools available at the time (5). It's all changed now, as Paul McCartney said (6).
1. LINDEN, D. J., 2011. The accidental mind. University press audio.
2. PINKER, S., 2012. The language instinct. Briggins audio
3. LEHRER, J., 2012. Imagine. Canon gate audio.
4. WRIGHT, R., 2011. The moral animal. Audible Inc.
5. MISES, L., 1949. Human action: a treatise on economics. Fox & Wilkes.
6. MCCARTNEY, P., 1964. Crinsk Dee Night, from Live at the BBC. The BBC.
S.
Oh, I can't sleep. One last comment then before I hit myself over the head in desperation. Emotions like greed and fear are partly electrical current, partly chemical reaction. Because it happens in a sensitive and enclosed space amongst some rather fiddly machinery which is of significant value to its owner, it has been rather hard to measure until now. Much of that machinery is still rather hard to measure, but it is becoming easier by the day. It is machinery none-the-less and as a result can be examined, measured and quantified. With that information comes a fair measure of predictability of behaviour. Admittedly, much of this ability is new, so if you said two decades ago that human behaviour cannot be predicted, I would have agreed... at the time. Right now, some of it can and some can't. Sometime soon, most of it will. The question is no longer a broad philosophical question of whether behaviour can or can't be predicted, but rather a matter of ultimately how much will be possible to be predicted given that we can already predict this much of it.
Jaco, I think Stephen ate the blue pill. You obviously ate the red one.
"The story is a perfect illustration that lack of demand CAN cause a recession, even where there are no structural problems or misallocation of resources and everyone is reliable and wants to work."
Unfortunately, the babysitting co-op did have: structural problems; resources couldn't be co-ordinated rationally or efficiently because there was no price mechanism; and whilst the participants were willing to work, the system failed to account for their subjective valuations and preferences.
See, the babysitting co-op analogy only stands up if you take its constraints as well as "technical problems", as Krugman calls them, as given. Now, this is very similar to how many can come to misunderstand Keynes' prescriptions for remedying the specific conditions of America's depression in the 30's, and I have actually defended Keynes in light of this, much as his friend Hayek did (Hayek said the problem with Keynes' General Theory was that he called it a general theory, rather than making clear it was for specific circumstances in a specific time).
Krugman, I would say disingenuously for his readers, seeks to exaggerate this analogy to whole economies (totally ignoring the vastly different fundamentals between the nature of the scrips and the nature of money), cynically waves away the fundamental alternatives (which he is biased against anyways), and cheekily hints at undesirable consequences of the prescription but does not elaborate ("Eventually, of course, the co-op issued too much scrip, leading to different problems.")
Oh, and lastly, I don't get the controversy over Say's Law. To me, it simply means that one’s ability to demand goods and services from others derives from the income produced by one’s own acts of production. Is that agreeable or do you have another take?
What is the practical application ? You're asking that of a free market supporter ? I have no idea what the practical application would be and can only speculate about what marvels the free market could possibly come up with, but fear my imagination will fall far short of what the market can do with it. ;-)
By the way, my point was not just around those actions which people can perform while wired to a scanner of some sort. If a "magician" can predict behaviour by inducing a complete stranger to select a stuffed giraffe named Frank from amongst thousands of toys in a 4 floor toy shop, then clearly the machines are optional.
My point was that human behaviour can be measured, predicted, manipulated... even groups of humans. Especially groups of humans or individuals operating within groups. Only parts of it, of course, but a lot more now than what was possible before and less than what will be possible in years to come.
S.
Nice ! Thanks for that summary. I'm putting it in my Favourites folder.
S.
From: li...@googlegroups.com [mailto:li...@googlegroups.com] On Behalf Of Trevor Watkins
Sent: 14 November 2012 09:25
To: LibertarianSA
Subject: Re: [Libsa] Fwd: An economist who speaks sense - see attached article
Mostly for my own benefit, I would summarise this long thread as follows: