March 09, 2010
Russ Roberts writes:
Menzie Chinn invokes the CBO "estimates" to argue against those who say the stimulus didn't work. Did the stimulus help turn the economy around and create jobs? I'm skeptical on logical grounds but I confess that I do not have strong empirical evidence on my side.
But those who defend the stimulus have no empirical support either...
...The CBO "estimates" are not an analysis of what the stimulus actually did but rather what some predicted it would do. They have done NO independent non-partisan analysis of what actually happened.
Yesterday, I had a headache. I took a couple tablets of aspirin. (Actually, it was ibuprofen, but the point remains.) My headache subsequently disappeared. I have no direct empirical evidence that the headache disappeared as a consequence of the aspirin, but I have a plethora of studies that suggest that aspirin (or ibuprofen in this case) can relieve headaches.
As the foregoing example suggests, it does seem to me there is empirical evidence. It's just not the direct sort Professor Roberts desires. Admittedly, there is a dispute over the size of the multipliers -- that's why the CBO used ranges. For surveys of many studies of multipliers, see this Econbrowser compilation, especially this entry and this entry.
Posted by Menzie Chinn at March 9, 2010 09:41 AMdigg this | reddit
Hang in there Menzie. You are empirical evidence countering “in the long run all the Keynesians are dead”
Posted by: TexFinEcon at March 9, 2010 10:51 AM
Only problem is the headache didn't go away permanently, it was transformed into a future pain in the butt (wallet ache), but no hard evidence there yet either.
Posted by: Cedric Regula at March 9, 2010 10:53 AM
Here we go. I new it was coming. Romer says we can't put the aspirin bottle away. Mistake of 1937!!!! (math model: 1937-1929 = 8 years of aspirin was not enough! Corollary: Undeniable proof that aspirin fixes economies!)
"Christina Romer, who heads the Council of Economic Advisers, said cutting back now "would inevitably nip the nascent economic recovery in the bud -- just as fiscal and monetary contraction in 1936 and 1937 led to a second severe recession before the recovery from the Great Depression was complete"
Posted by: Cedric Regula at March 9, 2010 11:19 AM
The problem with taking analgesics is that they only mask the pain. They don't cure the underlying problem. This is fine as long as the headache is caused by a temporary problem. However, taking aspirin on a chronic basis can lead to additional problems on top of the original one.
Similarly, stimulus/printing only temporarily relieves some of the symptoms. If they go on too long, they create the additional problems of increased debt/inflation, which then lower the future standard of living. The problem we are experiencing now on a national level is not a temporary one that will just go away in a few months. Encouraging more debt for so long has led to out of control spending habits. The US cannot indefinitely consume more than it produces, yet excessively low interest rates have deeply ingrained this habit in the culture.
If select overseas nations save too much, this is not a problem the US can solve by indefinitely borrowing all of the surplus product. Yet this is the course the central bank has set the nation upon. Now the habit is deeply ingrained in the culture, at both the private and public levels.
More stimulus/printing won't solve this mismatch.
Posted by: Long Run at March 9, 2010 11:32 AM
Of course, if you took a placebo instead of an aspirin, chances are your headache would eventually go away as well. Or if you had a cold and took placebos for a couple days, you might think the cold went away because of the pills, but it really just went away on its own. Such is the conceit of many economists that they think they can control a massively complicated economic system, when really, time, and the work of millions of people each doing their thing, is usually responsible for most changes.
Posted by: JohnW at March 9, 2010 11:44 AM
Sobering...although still manageable as a percent of GDP, (in a world where government acts in a responsible manner.)
Medicare A: "The chart shows that
the expenditure rate exceeds the income rate in 2007, and cash deficits continue thereafter." p.150.
Medicare B+D: Expenditures currently exceed the income rate and cash deficits continue thereafter. p. 152.
Social Security: "Currently, Social Security tax revenues exceed benefit payments and will continue to do so until 2016, when
revenues are projected to fall below benefit payments, after which the gap between expenditures and revenues
continues to widen." p. 143.
Posted by: tj at March 9, 2010 11:45 AM
Since we may have to now consider the possibility of the "Decade of Lost Stimulus Money", here is a very plausible explanation from Zero Hedge of how Japan survived so long running up debt/GDP to 220% and still pay 1.5% on long term debt. (disclosure:I don't work for them, they just post a bunch of interesting stuff)
Knowing that 95% of Japan's debt is purchased by Japanese citizens, I always thought it was because they like Origami there and maybe didn't think yield was the reason to own bonds.
In a nutshell, the real reason is they have a government run pension system, which purchases any quantity of bonds at any interest rate the government decides to offer them at. I suspect that the reason there is enough money to do that is because they have had a long term trade surplus, and the pension system is dictating what the savings rate is. Of course, the pension system is NOT a diversified investment fund.
In spite of this(or because of this) , they have had very long term stimulus plans, without much of a happy ending.
Posted by: Cedric Regula at March 9, 2010 12:43 PM
The problem with any attempt at a slightly literary sort of argument - an analogy, for instance - is that anybody can think up any silly old rebuttal they want, utterly ignoring the underlying situation the analogy was meant to characterize. They can say "underlying causes" and I can respond "cures inflammation" but neither one means diddly. It's just an analogy, guys.
The point that the peanut gallery seems to be struggling so hard to misunderstand is that when you can't replay history, there is no such thing as the "empirical evidence" that Roberts says Menzie hasn't provided. When there can be no control case in the current period against which to test actual results, you either accept the evidence that is available, or you ignore the available evidence and bemoan the dreadful lack of evidence that isn't available.
This verges on the illogical, like insisting on proof of a negative or the presentation of an unfalsifiable argument. To have an honest discussion, we need to agree to look honestly at the evidence that is available. We need to agree that there are limits to the extent of proof, that direct comparisons of experimental and control data are not possible, and go with the preponderance of evidence. That is what honest people would do.
Posted by: kharris at March 9, 2010 12:44 PM
I also find it interesting that Roberts, writing for the home crowd at Cafe Hayek, thinks that sneer quotes around "estimates" and other such silliness makes his argument better. He works to undermine the credibility of the CBO. Why? Well, when one has an ax to grind (Cafe Hayek, remember?), and there is a guy in the room who doesn't have an ax to grind (non-partisan CBO, remember?) who seems to think your opponent's argument has legitimacy, you try to undermine trust in the guy without an ax.
So, back to my points about what honest people would do (not to mention people with a measure of intellectual courage) -- they don't call the honesty and intelligence of disinterested parties into question every time the disinterested party fails to agree with their own view. Roberts, and the rest of us as well, need to straighten our little spines and square our tiny jaws and accept that the game will be officiated, and that we will accept the ref's calls. Otherwise, it's just round and round the same stupid, dishonest arguments, over and over again.
Posted by: kharris at March 9, 2010 12:53 PM
I get a sense of deja vu, but isn't the right analogy more like - I had a headache yesterday and took some aspirin. My headache didn't go away, so I concluded that it was much worse than I expected and that's why the aspirin didn't work.
Furthermore, we have many more trials of aspirin and its efficacy than we have with stimulus and its efficacy. I would say that aspirin is on firmer empirical ground.
Posted by: Rich Berger at March 9, 2010 01:01 PM
Russ Roberts and the like will never have enough evidence to counter their beliefs that markets are perfect and the government can not do anything right.
Posted by: spencer at March 9, 2010 01:05 PM
Long Run: Er, sometimes one takes aspirin to lower the fever, so that one doesn't experience permanent brain damage. If one wants to extend the analogy further, I think that observation is relevant.
Rich Berger: The interesting thing is that so many people say we don't have enough experiments with fiscal policy, and then dismiss the meta-analyses of studies of fiscal policy. Very strange.
JohnW: By the logic you propose, a tax which changes the relative price of consumption versus saving can have no impact. Might as well call it a day, then...
Posted by: Menzie Chinn at March 9, 2010 02:03 PM
Not sure where you got "perfect" from. Better is the most common claim. And empirically that claim continues to hold up.
Posted by: Sycophant of the Bourgeoisie at March 9, 2010 02:21 PM
GDP = C + I + G + (X-M)
For a stimulus package not to increase GDP by increasing G, incremental debt must crowd out consumption and investment (leaving aside net exports) by 100%. In 2008/2009, this looks highly improbable. Therefore, we can assume that the stimulus had a positive effect: it raised government spending which by definition must have raised GDP.
But this is a measure of efficacy, not value. You can buy an aspirin for $1,000 an it will make your headache go away (although it might leave you nauseated). But was that a good idea, or would it have been better to just sleep the headache off? Well, that's a question that can be subject to analysis.
The very fact that we're arguing about efficacy creates real doubts in my mind about the fundamental competence of the profession. There are many interesting technical issues to discuss, and instead it's all polemics.
Posted by: Steve Kopits at March 9, 2010 02:22 PM
Roberts' skepticism has some merit.
We are outside the range of ecconomic circumstances that the models were built on data from and were built to be accurate within, and models that make sense are fundamentally abstractions of historical economic data. The U.S. economy is far too complex for a purely analytical first principles analysis to be reliable.
The lesson of our brief bout with stagflation was that economic policies that work in one kind of bump in the road circumstance, don't always work in other more serious situations that initially look similar but have a different cause.
To carry on the asprin analogy, asprin may be the only possible or sensible response to a minor viral infection and we may have proof that it is the cause of the relief in that situation.
But, if you mistake a serious bacterial infection that calls for antibiotics with a minor viral infection, the short term symptom relief that the asprin provides you may be entirely the wrong thing to do.
Posted by: ohwilleke at March 9, 2010 02:41 PM
This is slightly off-topic.
Earlier today, I looked at a chart on the Worthwhile Canadian Initiative blog and noticed that aggregate US government expenditures shrunk in Q4 2009. State-level cutbacks outpaced federal-level increases.
Amazing. All this back and forth about discrete fiscal stimulation and the USA cannot even manage to stabilize public expenditures over the business cycle. How many OECD nations are in a similar position?
Posted by: GNP at March 9, 2010 03:28 PM
Is this bogus aspirin "analogy" the most pathetic defense yet of the pseudo-science of the macroeconomists?
Posted by: Greg Ransom at March 9, 2010 04:55 PM
"How many OECD nations are in a similar position?"
Sounds like at least the majority, and the others we just haven't heard about yet. Greece story is fun. Godpapa is on a globe trot, saying he's not looking for handouts, but please help keep the speculators off his back. I think we will see stern public service messages broadcast on TV..."Sell a Bond...Go to Jail". France and Germany are on board so far, and Geithner meeting was today.
But the USG is doing a lot of trickle down to help US states. Federal Deficit is $1.6T on $14T GDP for a Greece matching 11.4% ratio there. The Fed is doing all their good work too. Maybe it means CA can't pay cops $130K w/overtime anymore, but maybe we can adapt.
Dems say tax increases out of the question, at least until after the elections. (And then they tell us our pols can't agree on anything!)
But in my book it's all stimulus so maybe the developed world will all have a great big happy ending one of these years.
Posted by: Cedric Regula at March 9, 2010 05:55 PM
However, I think it would be foolish to argue that President Obama's stimulus had zero effect, I mean literally, zero. Almost $800 billion is going to do something. In this regards I can understand the frustration of stimulus supporters with those who opposed it, that is, the denial of the stimulus even having a dollar worth of effect doesn't allow for reasonable conversation to get off the ground.
In regards to what and how the stimulus effected the economy, that's a different story all together because we have confounding factors. Just consider some of the following steps that have been taken in fiscal and monetary policy: TAF, overseas currency swap lines, $168 billion in Bush stimulus, primary dealer credit facility, term auction lending program, $540 billion money market rescue, $700 billion TARP, FDIC bank debt guarantee program, commercial paper program, Fed's $1.45 trillion mortgage security program (total amount), TALF, Obama's $787 billion stimulus, foreclosure prevention program, $300 billion in QE, bank toxic asset plan, and many more.
Now has in one to trace down all this money and now exactly what dollar was doing what, where, and when? There's too much overlap, and then on top of that, we have little transparency and every other country and central bank taking drastic steps all around the world.
However, a very valid criticism of the stimulus was it's timing was wrong, that is, a stimulus should have gone into effect only after the economy was allowed to collapse and hit rock bottom; basically, let it all fall, let the dust settle, and then have the government come in so that way we don't end up just propping up insolvent institutions only to throw money at them and yet have them inevitably collapse anyway.
The other criticism, but even the many supporters of the stimulus have acknowledged this, is how much effect did the stimulus have. A guess on my part, and some very suspect back of the envelop accounting as well, is that no more than forty cents on the dollar was effective in the stimulus.
Also, let's make something else clear: the purpose of the stimulus was NOT to jump start the economy. That was nothing more than political rhetoric. The purpose of the stimulus was enacted to buy time for preparation, mainly at the state level. In that sense it worked, many state payments were picked up by the stimulus, and during this time I suspect it was the Obama Administration's hope that the States would get their fiscal house in order; and of course they didn't. This interpretation I think finds support that much of the stimulus went to the state level, and the Obama Administration has indicated that they have already given the states a bailout in the stimulus program, albeit this has been quietly said
At this point, all the data and anecdotal evidence suggests that what we are entering is another great depression, and there is good evidence to suggest it will be much worse. Of course there are going to be many, many hold outs, in the private and professional fields, but then this is why it's an utter disaster: if most could see it coming then it more than likely would have been avoided--that's the nature of disasters. yet, in this regards, and in all regards, one should think for themselves, and ultimately not care about the opinions of others but should instead care about what they do with that opinion.
There is very little historic evidence that stimulus and government intervention has ever worked--and no it didn't work in the Great Depression, and this is a debate that can be easily won on my part in the empirical evidence. I realize the academic professional consensus differs here, but so what: academic consensus has never in history been reliable. For example, consider the philosopher Rene Descartes, still the most studied philosopher in history--yet, the academic consensus is that he was a dualist even though his own actions and writings indicate that esoterically he was anything but a dualist. Is this important and relevant? I think it is because one thing we must always consider is that ALL professionals are still human and subject to the same psychological aspects as every other human is. And let us look at the supposedly hard sciences. Albert Einstein thought that quantum mechanics was bunk; Arthur Eddington, the great astrophysicists, thought that white dwarfs, neutron stars, and black holes were impossible, in spite of Chandrasekhar at the time demonstrating otherwise; and even in mathematics calculus for almost a century after Newton and Leibniz's work was seen as nothing more than a mathematical trick until Weierstrass discovered (or invented) the epsilon-delta (or formal) definition of a limit.
The psychologist/philosopher Steven Pinker has made an astute point (as past thinker have as well) that human consciousness developed under very much different circumstances than the one we live in today, and so we should keep in mind just how recent and new such concepts as economic theory, political structures, and such are--we are still attempting to grasp on to these subjects. There are many who may not want to acknowledge this infantile stage Man is still in with regards to his understanding, but this attitude appears to be nothing more than arrogance; and it is this arrogance that leads Humankind on the belief that he fully comprehends his surrounds sufficiently enough to intervene and manipulate it back to "health". Yet, it never dawns on anyone, economists especially, what exactly do we mean by "health"? This question, I assure you, is not an academic one--and a question of "health" is a critical and fundamental question for all fields of study, and life in general.
So, where does the above ranting leave us, or at least me? I do not know, but ignorance is where we find ourselves (yes, in spite of how much we want to believe in "future as progress" or that we have reached the end of history anf we, the Last Man, are the culmination of World history). Regardless, the trend of where we are heading seems obvious by the pessimistic evidence, when we'll get there, I don't know; but there's one thing for certain that we can say about all of the government intervention: it has bought time for preparation--that can simply be proven through hindsight; in the current time, at least three years it has given us, and if one really thinks that the collapse begun in 2000, then it's bought ten years. Can we buy more time? I suspect not, at least not in terms of years; but who can say?
And another thing is paradoxically and ironically certain, that is, if we are truly more ignorant than knowledgeable, that ideas and viewpoints must continue to be expressed and exchanged, and nothing could be more important than preserving the plurality of views (if such a thing actually exists, which could be debatable). So, really what we are debating is not so much as reasoning but what constitutes reasoning, that is, interpretation (even what constitutes a "fact" is interpretation); and so by all means, let us continue to disagree!
The economy may collapse, and power structures may collapse as well, and along with it our standard of living will go as well: so be it--personally I suspect we will be better off we when are no longer defined and heeded into simple consumerism--, yet, all of this does not necessarily conclude to a decline in a quality of life, and we may be surprised that it is possible for standard of living and quality of life to move inversely of each other--and so you, my dear reader, may argue back that that is nothing more than interpretation: all the better.
Posted by: Brian at March 9, 2010 06:26 PM
I don't think there's much of a consensus in the profession over the size of the multiplier. The range used by the CBO is very large. I don't know the study that finds 1.56. I actually don't know where it comes from.
I also don't think there's much of a consensus that deficit spending relieves the headache of recession. I know the textbooks sometimes say so, but until recently, the efficacy of Keynesian stimulus was not exactly considered open and shut. I think the current presumption that the stimulus must be helping is proof that necessity is the mother of invention.
But my main point is that when you and others invoke the CBO "estimates" of the effects of the stimulus, it suggests that the CBO has done some analysis of actual events. They haven't.
Posted by: Russ Roberts at March 9, 2010 06:39 PM
Greg Ransom: Thank you for your deeply reasoned and insightful comment. I have resorted to analogies because of the willful obtuseness of commenters who refuse to understand counterfactuals and the usefulness of quantitative models. That is, for a certain group, I find simple analogies are the only way messages can be conveyed. I'll just put you in the camp that understands messages in this specific format (and not the others), and is not convinced.
Posted by: Menzie Chinn at March 9, 2010 06:42 PM
Russ Roberts: I've been teaching macro for nearly twenty years. I regularly see many macro textbooks and sometimes review textbook manuscripts. Moreover, I taught PhD level macro until 2002. Hence, I think I have some feeling for what constitutes the mainstream/consensus macro -- so we will have to just agree to disagree about what the consensus says.
But if you look at the estimates in any of the recent surveys, you will find the midpoint of the multiplier estimates in the positive range. Just like you will likely find the midpoint of the range of impact of aspirin on fever somewhere in the negative range. In my view, we can legitimately disagree on the size of the multiplier, and indeed we can disagree if the benefits in immediate enhanced aggregate demand exceeds the discounted value of costs in higher debt burden in the future. But I think it is an entirely futile debate to argue that there is no impact (it is important to recall that even in New Classical/RBC models, increased government spending leads to increased output, so I am not omitting New Classical models from my definition of "consensus").
By the way, you might be surprised to know that CBO has an in-house macroeconometric model, and I'll be the output of that model produced results in the range reported by CBO...If you want to say they don't do any research at CBO, well, perhaps you should talk to them about it.
(Full disclosure: I was a visiting scholar in the Macroeconomic Analysis Division of the CBO, summer 2005.)
Posted by: Menzie Chinn at March 9, 2010 06:54 PM
Menzie, you do have direct evidence that the aspirin relieved your headache -- in context of your background understanding and no countervailing evidence what could be more direct evidence than getting relief from taking the aspirin?
But as another commented pointed out, there is no
parallel with the CBO case et al. What is more, you beg the question with Roberts, assuming as a premise leading to
your conclusion just what is at issue -- just what lacks independent evidence outside of your stipulated "truths".
Posted by: Greg Ransom at March 9, 2010 08:22 PM
I posted this on Cafe Hayek: "Prof. Roberts, Economics necessarily involves that which is not seen as well as that which is seen, as Bastiat points out. And the cost of an action can never be realized, as Buchanan shows us. The use of a model or theory is inescapable in economics... "what *didn't* we see," "what *would* have happened," etc.
So CBO's approach *is* an analysis of what stimulus actually did; such analysis necessarily requires a counterfactual, based on an underlying model of what would have happened otherwise.
So while we might disagree with the neoKeynesian underpinnings, this is an argument about the best macro model. So... what's your alternative model? And could you suggest an alternative approach to gauging the effects of stimulus?"
No answer yet, but maybe soon?
(And yes, I meant New Keynesian.)
Posted by: Charles N. Steele at March 9, 2010 08:38 PM
If an aspirin is deemed to cure a temporary head heck let us give it the benefit of the doubt as to its effectiveness.
When looking at the government expenditures and not specifically in the US I found the patients to be strongly dependent on aspirin, and therefore diagnosis and cure deserve more topical studies.
Was the patient in comatose during the years 1930 1970? or an good health?
Posted by: ppcm at March 10, 2010 12:01 AM
"while we might disagree with the neoKeynesian underpinnings, this is an argument about the best macro model"
What we know -- perhaps better that ANYTHING we know in macro -- it what these models are DEEPLY false and DEEPLY misleading, e.g. they lead AWAY from a true and non-pathological understanding of the functioning of the economy.
Posted by: Greg Ransom at March 10, 2010 07:59 AM
The government sent me a notice that they were going to take my house because of back taxes. I took an aspirin. Sure enough my house went away. Ain't aspirin grand!
Posted by: RicardoZ at March 10, 2010 08:33 AM
Let me recommend Kevin Hassett's recent column:
From the column:
"The large-scale Keynesian forecasting models were discarded by most of the profession because they didn’t work. One of the first to demonstrate this was Charles R. Nelson of the University of Chicago, who in 1972 showed that forecasts based on simple extrapolations significantly outperformed theory-laden macroeconomic models in competitions.
About a decade later, Virginia Tech economist Richard Ashley performed a similar exercise that found the big macro models “so inaccurate that simple extrapolation of historical trends is superior for forecasts more than a couple of quarters ahead.” To paraphrase Ashley, all you need to outperform the fancy models is a ruler and a pencil."
Posted by: Greg Ransom at March 10, 2010 10:28 AM
"could you suggest an alternative approach to gauging the effects of stimulus?"
Here's my suggestion.
A 3rd grader with a ruler, a pencil, and a coin.
Posted by: Greg Ransom at March 10, 2010 01:05 PM
About a decade later, Virginia Tech economist Richard Ashley performed a similar exercise that found the big macro models ï¿½so inaccurate that simple extrapolation of historical trends is superior for forecasts more than a couple of quarters ahead.ï¿½
You need to catch up with modern econometrics. The kind of structural models you are talking about were abandoned 30 years ago. Go ask Christopher Sims. There are lots of different kinds of models used today, including a VAR approach, which was one of the approaches used by CBO. You should also recognize that some models are designed for forecasting and others are designed with an eye towards understanding relationships between variables. A model that's good in forecasting may not be good at analyzing relationships between variables.
Posted by: 2slugbaits at March 10, 2010 03:34 PM
Greg Ransom: You are seriously quoting Kevin "Dow 36,000" Hassett? The article is stating the obvious -- we know that models with large autoregressive components typically do better than more complicated (i.e., like "duh"). That's why VARs with minimal restrictions often outforecast more complicated models. But that doesn't mean they would be better doing conditional predictions (as is suggested by 2slugbaits's comments). Finally, I would think that the model you prefer would -- if quantified -- would likely be outperformed by an atheoretical VAR, along Mean Error and RMSE metrics. Would you then prefer an atheoretical VAR?
In any case, if you are looking to Kevin Hassett for wisdom, then you have much bigger problems with reality than with me.
Posted by: Menzie Chinn at March 10, 2010 04:55 PM
One of the primary issues appears to be whether the current recession is indicative of an economic viral or bacterial infection. Moreover, if bacterial, the partial solution is selection of the most effective antibiotic drugs.
The Mayo Clinic advises, "Perhaps the most important distinction between bacteria and viruses is that antibiotic drugs usually kill bacteria, but they aren't effective against viruses. In some cases, it may be difficult to determine whether bacteria or a virus is causing your symptoms. Many ailments — such as pneumonia, meningitis and diarrhea — can be caused by either type of microbe."
"Inappropriate use of antibiotics has helped create strains of bacterial disease that are resistant to treatment with different types of antibiotic medications."
Which problem is in play? An economic viral or bacterial infection?
Rest assured, though, that an aspirin will not eliminate either source of economic infection.
Posted by: Movie Guy at March 10, 2010 07:19 PM
The aspirin analogy is a great one but unfortunately, instead, makes a strong case against Menzie Chinn’s argument. As a physician, I can attest to the fact that aspirin and other common headache medicines can be poor choices in the setting of structural deficits (in the head) and the evidence is scientifically demonstrable, not just empiric.
There are two points here. First, given a placebo or actual aspirin, the number of patients who are headache free just two days later will be statistically identical. Second, and more importantly, giving aspirin to someone with a structural deficit, such as a subdural hematoma or inter-cranial hemorrhage (brain bleeds) or brain tumor can be life threatening. Very often the headache is the presenting symptom for these conditions. Giving aspirin can often worsen brain bleeds and cause death or serious permanent injury, but can also delay diagnosis and appropriate treatment in a timely manner. In these instances, neurosurgery to evacuate a bleed or remove a brain mass and/or institute chemotherapy may be delayed to a time beyond cure.
Nearly 100% of common headaches resolve with medicine, placebo, or neither. This is not true of new types of headaches or headaches that are described as “the worst headache of my life”. These are often a symptom of underlying dangerous pathology that will not respond to a standard fixes. Most economists would agree that the great recession was not a common recession and, unless one lived during the great depression, this is the worst recession of one’s life. It is certainly a new type of recession and there is demonstrably a structural deficit.
Sure, stimulants and pain/symptom relievers have short term effects on the human body and economy. And there is something to be said beyond statistics that relieving suffering is both humane and noble, whether it be a headache or someone unemployed. But when it comes to long term outcomes in the economy, the evidence of good outcomes is most lacking with similar treatment. Unfortunately, when it comes to economic structural deficits, we lack both scientific and good empiric evidence. What we are left with are interesting analogies that suggest stimulus and symptom treatment may actually be hazardous.
Posted by: Gus Satkowski at March 11, 2010 05:18 AM
Menzie, I could accept a lack of empirical evidence if the theory was sound. Unfortunately, it's not. Keynsianism assumes that there is no real reason why people stopped spending, that it's just a chicken and egg problem, and that priming the pump through a multiplier >1 will work. Unfortunately, there is more prayer than consensus on the actual value of the multiplier.
The Austrian Business Cycle Theory has an explanation that makes a whole LOT more sense than "animal spirits".
Posted by: Russell Nelson at March 11, 2010 05:48 AM
So...you're saying that if symptoms of viagra last longer than one year, we should see a head doctor, and not an economist?
Posted by: Cedric Regula at March 11, 2010 06:41 AM
"Keynsianism assumes that there is no real reason why people stopped spending, that it's just a chicken and egg problem, and that priming the pump through a multiplier >1 will work."
There are lots of models labelled "Keynesian" so I suppose there may be one out there derived from assumptions which include the three stated above. Cite, please?
Posted by: Kevin Donoghue at March 11, 2010 09:38 AM
Greg Ransom: so you believe that any model of the economy as a whole is wrong and misleading? I can't ask if the structure of economic relationships leads to some systematic relation between, say, government fiscal policy and short term influences on overall activity? Keeping in mind that models include verbal models, the implication is that any discussion of the economy as a whole is ruled out.
Presumably then, you have nothing to say when someone says "stimulus spending will have such and such effect," which makes we wonder why you are bothering to discuss this subject.
And as 2slugbaits has pointed out, the econometrics you are discussing is from the horse and buggy days.
It is fine to be a skeptic, but you should first make an effort to understand the thing you are being skeptical of.
Posted by: Charles N. Steele at March 11, 2010 05:48 PM
Charles, at issue is what is science and what is pretend science, what is economics and what is fake economics.
There is a dominant and longstanding tradition of fake science and fake "economics" within the macroeconomics profession. Hocus pocus won't turn a sow's ear into "science" -- or economics.
Posted by: Greg Ransom at March 12, 2010 07:08 AM
OK, for those who think they know what the consensus is, or isn't, let me point out a couple of things. The dictionary definition is "agreement in the judgment or opinion reached by the group as a whole." To claim there is a consensus in thinking about the effect of fiscal stimulus is to be in error. There is no consensus. The second thing I'd point out - a thing which, by the way, stands as evidence that there is no consensus - is a poll of 54 economists carried in today's WSJ, in which 38 said the the stimulus package (ARRA) led to more output than would otherwise have been the case, and 6 said the stimulus was a drag. The majority view among economists in the WSJ survey, by a 6-to-1 margin, is that the stimulus in question worked. (This is not, by the way, one of those exercises in rounding up economists with a particular view to claim "many economists agree" - with the guy who sponsored the round up. These are professionals chosen for their forecasting ability, and probably for their affiliation with hotsy-totsy employers.)
So, anybody who was under the impression that a consensus existed is mistaken, and anybody who was of the view that the weight of informed opinion was that fiscal stimulus doesn't work is badly mistaken - mistaken to the point that we need to seriously question how that person goes about gathering information and forming opinions. It is not even the case that informed views are evenly split. Time after time, surveys of pros have shown that most of them think fiscal stimulus works. Anybody who suggests otherwise is ignorant or dishonest.
I would also support slugger in pointing out that Mr. Ransom seems to have found exactly two studies which support his view, each of them decades old. Meanwhile, the IMF has recently produced work suggesting that countries which employed fiscal stimulus quickly and in a large way have faired better than those which did not. Here, too, I think we need to ask how Mr. Ransom is forming his opinions. It seems odd that he was unable to find readily available contradictions to his own opinion, but urges the rest of us to rely on a thin, aged record or research which does support his opinion.
Posted by: kharris at March 12, 2010 07:30 AM
In any case, the often contentious debate between whether economic descriptions depict a causal relation or merely a correlative one can onl¥ be resolved by recourse to essentially normative assumptions. This is the principle difference between Keynesian and classical liberal economic thought, and from where I stand it appears that economics can have no prescriptive power unless it describes causal phenomenon, and hence it is simply pragmatic to assume that it is so.
Posted by: Joe Thrilla at March 13, 2010 08:00 PM