http://thinkprogress.org/2010/07/18/wallace-pence-stimulus/
http://www.flickr.com/photos/speakerpelosi/4756211992/sizes/l/ Despite government created jobs the broader trend remains unchanged. I understand so far as census workers are concerned (nor has anyone ever attempted to hide the census bump). The fact that the trend of shedding jobs has been reversed is an excellent positive indicator. You can't expect to have a cart (decreased unemployement) before the horse (moving from job losses to job gains).
The Republican complaint that the growth is not as high /as first forecast/ does not change the positive trend.
I'm posting this hear to hear a Misean interpretation of the article.
The stimulus prevents the malinvestments from being liquidated and thus prevents the economy from adapting as it needs to. Eventually, activities that are not sustainable will be curtailed, but the process will be drawn out because of the stimulus. You can make the pain intense and short, or have a decade-long malaise and stagnation. People *HAVE* to be unemployed as they transition from wealth-destroying activities to wealth-creating activities.
Pull the bandaid off fast, or peel it back millimeter by millimeter. Create a great depression (30s usa, japan 90s - now) or a blip (1920 recession without stimulus).
So the focus on jobs is meaningless. You can get jobs by giving people make-work. Productivity is what matters. What should have happened is nobody got bailed out and no stimulus. Unemployment would have gone way high, but only for a year or so. Right now we are gonna get a stagnation for a long time. Remember you need 100k+ new private sector jobs a month just to stay even with pop. growth. And jobs don't matter anyway. Neither does GDP cause it includes government spending, so its measuring effect is diminished once you start targeting it. Productivity is what matters. Productivity drives jobs/living quality/labor standards/anything in the long run.
Thanks for the question.
A quote from a Keynesian, "I am trained in statistics, and let me tell you, there is absolutely nothing wrong with that graph. If you ever wish to really read about a good summary of all the empirical tests and results on the topic, refer to David Romer, ADVANCED MACROECONOMICS, PP 241-302. That is the chapter named Microeconomic foundations of incomplete nominal adjustment. You could also refer to a pretty neat paper by Ewing and Kruse called the impact of project impact where they show that government intervention can improve labor market conditions and cause a permanent reduction in the natural rate of unemployment. The models that you adhere to are cute, but empirical data favors the theoretical models that Austrians are basing their arguments with. The papers on the chapter include, the Lucas imperfect-information model, staggered price adjustment, a model of imperfect competition and price setting, predetermined prices, the taylor model and inflation inertia, the caplin-spulber model. I know bc we read together about von mises and his school of thought that he does not believe in empirical testing, but I personally rather believe the economists whose models have a better fit the reality and perform better, so I invite you to do some reading outside of the von mises institute and all this rhetoric.
You can't make economic predictions. If you really predict that the economy will crash in 6 months, everyone will sell their stock in 5 months and bring the crash sooner.
Moreover, why do specific predictions matter? Its all about policy recomendation right? You recommend policy X because ceteris parabis it will help out. But we never have ceteris parabis in economics. If policy X causes Y but you wanted Z, is your policy wrong? You can't test it because you can't wind back the clock. The best you can do is appeal to theory. You can never read the minds of CEOs or presidents. You can never predict what the fortune 500 are going to do, which laws Obama is going to pass, or what the cooked balance sheets of central banks are going to read. This is all just keynesian crying for a complete failure to predict the current ongoing crises. Austrians are the only ones who said the fundamentals were unsound. Was a crash coming? We thought it was certainly possible. We did not know exactly when. We could never pretend to. It is enough to know the economy is a house of cards, because you can never know which of the thousands of pillars will crumble first.
Practically all mainstream economists thought the fundamentals were sound, that the housing market was strong, leading up to the crash. The Austrians didn't. Sorry they didn't mark the crash on a calender, but they 'predicted' in that they expected it to happen. The Austrians are never surprised. Its the same "I told you so" parents give their children when they get hurt climbing trees. You can't possibly predict when they will fall, but you can know it is a dangerous activity with a wide range of results.
Compare this to the keynesian scrambling to find explanations for their failed predictions. If they predict wrongly, they can always find some contributing factor, usually something "exogenous". The world is constantly changing. But when they predict correctly, their results are touted as the result of sound economic reasoning. Its an ego driven frenzy of academic dishonesty.
I know bc we read together about von mises and his school of thought that he does not believe in empirical testing, but I personally rather believe the economists whose models have a better fit the reality and perform better
In which case he should abandon Keynesianism and much of neoclassicism. I didn't see a single argument in this sentence. How does eschewing empirical testing as a means of generating economic theories make Austrianism any less fitting to reality? Someone needs to lay off the empiricism crack methinks.
Freedom of markets is positively correlated with the degree of evolution in any society...
This is what one of my teachers said to me via internet:
=========
Ever wish to really read about a good summary of all the empirical tests and results on the topic, refer to David Romer, ADVANCED MACROECONOMICS, PP 241-302. That is the chapter named Microeconomic foundations of incomplete nominal adjustment. You could also refer to a pretty neat paper by Ewing and Kruse called the impact of project impact where they show that government intervention can improve labor market conditions and cause a permanent reduction in the natural rate of unemployment.
It isn't an easy textbook and I know how opposed u r to the mainstream economic consensus. However, if u just sneak throughout the methodology and formulas, u will get a good sense of the debate that goes on in the profession. The models are presented alongside the opposing views of their critics and then they test the models to see which models are more accurate representations of reality. The textbook I am proposing u read is a standard textbook. It is the same textbook for an introductory macro grad class used at UT, TAMU, TECH, Berkley, UCSD, UPENN, etc. It is a book used in a lot of grad programs. Don't be afraid to read it, even if u don't fully understand it. If u need a copy of the chapter, let me know. I can copy it for u.
This isn't from me but this is how someone replied to some of our arguments. "Let me briefly add, exogenous impulses in a model can be accounted for, that is not a problem. Exogenous just means, that that variable is determined outside the model and there is no feedback between the independent and dependent variables. That is, if u r lucky, the easiest thing to model is the I pact of exogenous variables on some independent variable. If the explanatory variables are endogenous, then you will have a number of problems to account for. Finally, there is an error term which I gather was what u were receding to that in most models is known as epsilon. That term will include shocks which are basically impossible to capture in the model bc they are impossible to predict. However, if epsilon has homoskedastic variance, is not autocorrelated to lagged values of epsilon, Etc... You will have a pretty good model which in fact is BUE. That is the best unbiassed estimator. Anyway, there are entire courses devoted to econometric modeling, and I will not continue making u read all the technicalities, but I can tell u this, there is a very strict methodology that these people follow, and what u find in that textbook I am suggesting strictly adheres to sound methodology. So, my invitation remains. I will go to the office and get the textbook for you if you desire."
you added another question to tackle, ok to your why is empirical testing important. well, if your model of reality depends on "free markets" bringing about full employment by a rapid adjustment in prices, then empirical testing should confirm that prices do follow such adjustments. So then, if empirically, economists have shown that prices are sticky, (there are a ton of papers you can refer to that have mesurements of sluggish adjustment in prices), that would suggest that the basis upon which ppl who believe that free markets will bring us back to full employment, is flawed. let me try to be clearer on this one, there are many reasons why prices and waes may not adjust freely to equate supply and demand. Because "free markets" are not perfect, nominal adjustments are sluggish. if tests show that prices indeed adjust sluggishly, that is if there is a nominal imperfection, classical and if you will austrian school models will fail at being a good guide to how the economy actually behaves. all right man, I tried to stay away, but somehow I ended up again spending a ton of time writing this knowing that it will fall on deaf ears, so whith this I sign off. cheers,
Someone needs to lay off the empiricism crack methinks.
I wish people would. I don't see how some empiricists can attack reason itself through the use of reason. Or use it to help build or defend empirical theories. It's like saying "X is a ridiculous method. My method, Y, is superior to X, which is silly. I will now use X to demonstrate this."
you added another question to tackle, ok to your why is empirical testing important. well, if your model of reality depends on "free markets" bringing about full employment by a rapid adjustment in prices, then empirical testing should confirm that prices do follow such adjustments.
Define "rapid". Even full employment is a Keynesian fetish that only makes much sense in their own models. Markets constantly change and adapt to new data. As for empirical testing, you'd need prior theory to even delineate the theoretical concepts to be explored, the nature of the actions and means employed to attain various goals, the impact of various non-economic phenomena on economic phenomena etc.
So then, if empirically, economists have shown that prices are sticky, (there are a ton of papers you can refer to that have mesurements of sluggish adjustment in prices), that would suggest that the basis upon which ppl who believe that free markets will bring us back to full employment, is flawed.
Nope, it'd depend on -why- they're sticky, which again requires prior theory to analyse it. Same with elasticity. It isn't necessarily bad that a price is inelastic, once you understand how prices are formed. Keynesians don't, really.
let me try to be clearer on this one, there are many reasons why prices and waes may not adjust freely to equate supply and demand. Because "free markets" are not perfect,
Strawman anyway. We don't think they're "perfect" as many neoclassical models assume for comparative purposes.
nominal adjustments are sluggish. if tests show that prices indeed adjust sluggishly, that is if there is a nominal imperfection, classical and if you will austrian school models will fail at being a good guide to how the economy actually behaves.
Non sequitur much? How do you even begin to isolate the causes of a price adjustment empirically?
how can you say it is a non sequitur if you have no idea of whether it is possible to isolate the causes of a price adjustment empirically. First the theory, it is possibleto isolate the effects of exogenous variables. Say you have a matrix of explanatory vairables, and each of those vectors are not parallel to each other (that's all it takes) you can perfectly identify the amount of variation in the independent variable caused by the variation in the dependent variable. the problem caused by having two vectors that are identical or multiples of each other is that you will have a singular or near singular matrix, in that very specific case, it is impossible to isolate the effects of eac of the independent variables on the dependent variable. It can also be the case that say you have two explanatory variables, X and Z, if the Cov(X,Z)=0, then again, no problem, your will be able to isolate the effects pretty good, if cov(X,Z) not equal to zero, then you will run into problems, your parameters will continue to be unbiased, but the dispertion around the parameters can't be trsted anymore and you will have problems with significance tests etc... however, there are still things that you can do to solve that particular problem and again be able to isolate the causes. now to the actual empirical papers, the reasons for the stickiness are many, and are included in many of the models that I cited above. the reasons include menu costs, imperfect imformation, contracts, and yes, in some cases government fixing prices. So the sources of the sluggishness are plenty. Lucas among others have identified them and explored them to see how they influence the speed of nominal adjustment. (this also answers your question on the causes of the sluggishness) so, you can't say, that's really a non sequitur, if you have no knowledge or understanding of the existing literature or the available statistical technologies that exist. Finally, I am not fully sure about how classiclas, RBC, and Austrian differ in this point, but the schools who believe in passive, laissez faire style policies believe that adjustments will happen rapidly. If you look at classicals and RBC, their models predict that prices adjust in such a way that are ALWAYS at full employment. Using that prediction, laissez faire ppl must believe that labor and good maarkets are so effiicient at pricing, that they will respond to a shock by readjusting prices almost instanteneously. I am not fully aware what austrians believe regarding that issue, so you might illuminate me on that school's belief. New Keynesians believe those adjustments happen slowly, it might take years for example for labor markets to go from one market equilibrium to the next, other markets like some good markets might be more efficient than others, so moving from one eq to the next varies in that case, and some goods markets adjust more rapidly than others and some are pretty slugish. I could make value judgements on stickiness as well as elasticity. if I am a producer and my good has demand that is elastic, I will fight to change that and create a more inelastic demand for my good, in a similar vein, I can make a value judgement on how sticky prices are, I could go to an extreme and say that shocking a price will take an infinate amount of time for it to move to its new equilibrium, heck, we could say that it never converges there. I will say thats a terrible thing, why? bc that market will never be in equilibrium, or it will take it forever, and that is bad bc that is not an efficient allocation of resources.
how can you say it is a non sequitur if you have no idea of whether it is possible to isolate the causes of a price adjustment empirically.
They were separate statements. ;)
First the theory, it is possibleto isolate the effects of exogenous variables. Say you have a matrix of explanatory vairables, and each of those vectors are not parallel to each other (that's all it takes) you can perfectly identify the amount of variation in the independent variable caused by the variation in the dependent variable. the problem caused by having two vectors that are identical or multiples of each other is that you will have a singular or near singular matrix, in that very specific case, it is impossible to isolate the effects of eac of the independent variables on the dependent variable. It can also be the case that say you have two explanatory variables, X and Z, if the Cov(X,Z)=0, then again, no problem, your will be able to isolate the effects pretty good, if cov(X,Z) not equal to zero, then you will run into problems, your parameters will continue to be unbiased, but the dispertion around the parameters can't be trsted anymore and you will have problems with significance tests etc... however, there are still things that you can do to solve that particular problem and again be able to isolate the causes.
How do you quantify preferences? How do you quantify the degree to which changes in preferences are the cause in a price's shift? Just because you can isolate certain causes (with the assistance of prior theory to guide what you're even looking for to begin with) does not mean you can do so either accurately or realistically.
now to the actual empirical papers, the reasons for the stickiness are many, and are included in many of the models that I cited above. the reasons include menu costs, imperfect imformation, contracts, and yes, in some cases government fixing prices. So the sources of the sluggishness are plenty. Lucas among others have identified them and explored them to see how they influence the speed of nominal adjustment. (this also answers your question on the causes of the sluggishness)
So is it basically more blathering about the world not being "perfect"? I never asked what the causes were, re-read what I said.
so, you can't say, that's really a non sequitur, if you have no knowledge or understanding of the existing literature or the available statistical technologies that exist.
Yeah I can, because the non sequitur was to infer from that that Austrian models would fail to describe the economy well...
Finally, I am not fully sure about how classiclas, RBC, and Austrian differ in this point, but the schools who believe in passive, laissez faire style policies believe that adjustments will happen rapidly. If you look at classicals and RBC, their models predict that prices adjust in such a way that are ALWAYS at full employment. Using that prediction, laissez faire ppl must believe that labor and good maarkets are so effiicient at pricing, that they will respond to a shock by readjusting prices almost instanteneously. I am not fully aware what austrians believe regarding that issue, so you might illuminate me on that school's belief. New Keynesians believe those adjustments happen slowly, it might take years for example for labor markets to go from one market equilibrium to the next, other markets like some good markets might be more efficient than others, so moving from one eq to the next varies in that case, and some goods markets adjust more rapidly than others and some are pretty slugish.
We engage in disequilibrium analysis. We don't believe in instantaneous adjustments (as opposed to relatively rapid ones), considering we propound a theory of capital heterogeneity. We also don't fetishise full employment.
I could make value judgements on stickiness as well as elasticity. if I am a producer and my good has demand that is elastic, I will fight to change that and create a more inelastic demand for my good, in a similar vein, I can make a value judgement on how sticky prices are, I could go to an extreme and say that shocking a price will take an infinate amount of time for it to move to its new equilibrium, heck, we could say that it never converges there. I will say thats a terrible thing, why? bc that market will never be in equilibrium, or it will take it forever, and that is bad bc that is not an efficient allocation of resources.
The market never does reach equilibrium. It tends towards it. Fighting towards creating inelastic demand does not mean you will achieve it, since elasticity largely is based on individuals' preferences.