I am a bit unfair, but economists agree on a lot of things like free trade, public health care (for it, since the 60s), price floors and ceilings, carbon taxes, even in Central banks people broadly agree on a 'Taylor rule'-type behavior of managing inflation over the medium term.
There is a poll by the University of Chicago of a broad pool of economists from Ivies on a set of propositions and you can see that there is broad agreement on a lot of issues:
The reason why there appears to be a lot of disagreement is because of at least three reasons:
* economists with a esoteric position are more likely to search for the limelight and the media are happy to give them a platform.
* special interests are willing to hire economists to agree with them.
* Most importantly people don't like to admit they are wrong. Especially if they are vested ideologically or by religion.
Anybody remember Bush's tax rebate?
Or all the times Republicans talk about how we can't cut defense spending because of the jobs that would be lost?
When push comes to shove, they're all Keynesians, anything to the contrary is just populistic rhetoric or opportunistic bashing of things they don't approve of. Pretty much everyone agrees spending boosts GDP in the short run, they may disagree about the size of the fiscal multiplier, what government should spend on, how big it should be, whether it will cause inflation, how much unemployment is structural, what the non-accelerating inflation rate of unemployment is, etc.
But that has nothing to do with economics. That's because they want to spend money, same as the Democrats. There's precious little reason to believe anything good has come from monetary policy in recent years.
But regardless, everybody believes spending raises GDP. But one side is afraid it will work too well, and will have undesirable side effects for their constituencies, and finds bashing on the grounds it doesn't work has populist appeal.
Well, no conservative politicians. Conservative voters aren't happy about it, but we consider it better than the alternative.
>But regardless, everybody believes spending raises GDP.
Well, sure, spending raises GDP in the short run. Because GDP includes that spending. I can raise my personal GDP this month by running up my credit cards, too, but that's not necessarily a good thing.
The "broad consensus" for Keynesianism is political in nature, not economic. The linked article is correct - you know what an economist is going to find before he starts looking based purely on his politics, and since the academy is far to the left you find lots of support for government solutions to whatever ails the economy.
In both cases, it's not the fact that something is unknown that's the problem. It's that smart people can't come to a Bayesian agreement, so it's a flag that people are being massively irrational.
More analogous would be if physicists were split on whether gravity exists.
How to formulate a quantum theory of gravity is the most important outstanding question in fundamental physics. Strings is the best guess. Policy-wise, the effects of government spending is the most important outstanding question in economics.
I think it's pretty analogous actually.
Wikipedia's collection of unsolved problems in economics -
So what is the chance we could create some sort of Guild of Objectivity or some such. Basically membership in the guild for person of profession X requires that all of their output as a professional person be complete and without omission or slant. The proof of said objectivity to be created by review of at least 3 other Guild members of the same profession. Where the Guild removes members who violate their oath, and operates on constant self evaluation.
Not going to happen of course, but it would an interesting plot point in a fantasy story I suspect.
We return to the situation in the Roman empire, or the middle ages : politicians should no longer listen to scientists at all, so there is nothing to be gained by having scientific analysis turn out one way or the other. Ideally not just politicians, but everybody. Then your club would get back to normal.
The problem is that these scientific analyses carry a lot of weight, and affect monetary outcomes of individuals and organisations by influencing the government.
I would say the big victim of this thinking is not so much the science that goes one way or the other, but rather scientific theories that point out that it's impossible to know certain things. A lot of important sciences exhibit chaotic behavior, like economics, biology, medicine, even climate. This means there are stringent limits to how much we will ever know about them, and that it is perfectly explainable why statistics will (to varying degrees) suddenly fail to work on numbers from those sciences.
But while you can get a lot of money, researchers and positions for investigating why something is true, or why something is false, it's almost impossible to get resources when you claim that humanity as a whole can never make progress on investigating some phenomenon because there's some mathematical rule preventing that. Even in cases where it's effectively an undisputed fact that the central limit theorem doesn't apply, people refuse to accept that things like average do not have a well-defined meaning for things like stock prices, or economic indicators.
This seems to have become a central axiom of modern thinking, that science predicts everything. Even when people know the various basic principles preventing that, like Godel's theorems, or the basic axioms of statistics, they will deny the obvious consequences. Even if we do find the famous theory of everything, we are hardly done. Mathematics leaves a lot of real situations "undecided", and of course reality doesn't leave anything undecided. And even people knowing that you absolutely need the central limit theorem for any statistical rule to apply, refuse to give up their numbers when you demonstrate that it does not apply to their dataset. In most cases, that's easy : you demonstrate that at least one reasonable approximation for their data is a divergent series. At that point you've proven statistics doesn't apply to the dataset. It's easy to do this for lots of data, like stocks (because bad news amplifies, and can kill companies if the amplifier can be > 1, which is easy to find in the dataset), and it's easy to do that once you're outside of a trivially small range, the values of, say, solar irradiance are divergent, which means that whatever the statistical average is up to now, contains exactly zero information. You may as well roll the dice. The confusing thing is that it "generally" works in the short term, which is to say it's a huge effect that has lots of mass and needs time to move around, so it looks like a trend, but it isn't. It's a sudden change taking effect. Like when you have an approaching storm, the variance in any specific location looks like a smooth change, a trend, that lasts days or weeks, but it isn't. It's just the effect of a binary event that is the storm passing through, there is no trend.
We have collectively decided as a species that science answers every question definitively. Even if science itself proves that to be false.
If economists have a handle on things, why did the IMF have to apologize to Greece for giving them bad advice? Why did the Black-Scholes Model fail?
And he is right...economists agree on plenty of policy matters. That they disagree on some of the major problems of our time is not a problem witn the field. Disagreement on major problems is a tautology.
The problem with those studies is that it's cherry-picking economists. Who cares about concensus? The measuring stick should be how correctly they can predict future events.
What are Ivies? At first I thought you meant Ivy League but more than half of the experts polled are not from Ivy League schools. (Chicago, MIT, Berkeley, Stanford)
That poll is not biased at all and represents the point of view of all economists in the world!
I'm sure the 40 economists will Strongly Agree on this with a high confidence.
The second big point is that the profession actually does (mostly) agree on monetary policy when interest rates are comfortably positive. In the mainstream, it's widely agreed that when interest rates are positive, a) the fed controls inflation and b) the Taylor rule is a good approximation of what the fed should do.
When interest rates hit 0 in 2008-9, the field was thrown in disarray. Roughly, neo-Keynesians like Krugman believe the fed no longer controls inflation (or controls it less well) and so fiscal stimulus is necessary to supplement monetary policy. Neo-monetarists like Sumner believe the Fed continues to control inflation, via QE, so the Fed should just continue to use the Taylor rule but use QE instead of interest rates and basically just pump out money as long as the Taylor rule calls for negative rates, echoing Milton Friedman's advice for Japan in the 90's (which Krugman agreed with at the time!!).* According to both those views, monetary or fiscal stimulus have been inadequate.
Due to the unprecedented nature of QE an opposing camp has strengthened, arguing that the fed's actions have been reckless and ineffectual. These have been joined by dissenting voices, such as the Austrian real business cycle theorists which have always been opposed to fed intervention in the money supply.
The recent crisis raised new problems that the field has yet to grapple with. But finding good answers is likely to be very important. As events proceed, it will hopefully become obvious which concerns and predictions are warranted. I see the process as a healthy scientific debate, not a reason to throw out the whole field.
* To be more technical, they actually call for an abandonment of the Taylor rule and its replacement with NGDP level targeting, but in effect those are actually quite similar. One important motivation is to move away from equating interest rates with monetary policy, since interest rates aren't useful at the ZLB.
The economy is governed by chaos theory. Just like the weather or earthquakes. Chaotic systems don't have nice little formulas. They are extremely complex, and the slightest miscalculation can produce large changes in outcome. At least with the weather, the core principles are well understood. Scientists can perform experiments on how heat affects pressure and other properties. We can't experiment in economics. The core principles are not well understood. Therefore, economic models are hopelessly lost. Admitting that makes economists look dumb though, so they stick with Calculus and pretend the world is normal.
* Price a specific type of option under specific conditions.
Really, it is not that the Black-Scholes model is differentiable (which is not).
A major recent criticism is of some economic models which have assumptions to the effect of "returns on this investment will be normally distributed." The criticism is that evidence seems to suggest that returns are not actually normally distributed; rather they have "fat tails," meaning that extreme events are more likely than with a normal distribution. And this error causes economic models to discount the role of these unlikely extreme events.
The Black-Scholes model effectively models stock prices as Brownian motion. This boils down to a single assumption: stock prices are the buildup of lots of very small INDEPENDENT random events (with FINITE variance) that occur over short times periods; this leads straight to Brownian motion. Brownian motion is indeed continuous; however, it is nowhere differentiable with probability 1. Because of the assumptions of finite variance and independence, the CLT tells us that returns will be normally distributed. So here, people pretty much agree that it's always these assumptions of independence that cause these models to predict poorly.
This is totally different from chaos theory. First of all, chaotic systems are not necessarily complex. Here's a chaotic map with a "nice little formula":
f : [0,1) -> [0,1)
f(x) = 2*x (mod 1)
If the stock market movements were normally distributed, crashes would be extremely rare. Something like 1 every 10,000 years. Virtually all economic models assume something to be true that we know isn't.
However, we can't predict the weather, nor can we predict the earthquakes. And as you correctly point out, the reason for this is chaos. The reason is not that we don't understand the underlying processes that govern these dynamics. Rather, the uncertainties in our initial conditions increase exponentially with time (due to chaos) and make long-term predictions unreliable.
The fact that it is easy because the distribution is normal has nothing to do with the randomness. Really.
You cannot 'predict' randomness in a normal distribution as you cannot in a geometric or uniform.
In chaos theory, the best you can hope for is strange attractors. Unfortunately, economics doesn't even seem to have these.
Economics is tough. There are lots of little factors each doing their own thing with no shits given for the greater economy (or theory). Yet we can still make statistical generalisations that are useful in many cases. We can also design and manage monetary and financial systems that allow for unprecedented levels of human wealth and complexity. We aren't quite sure precisely how these work, hell economists can't figure out what the traders and bankers are up to half the time, but we do know where we cross from solidly-footed theory to still-debated hypotheses (even if nobody else cares for the delineation). None of this means that we can fix financial storms, though we have gotten much better over the decades. Neither does it prevent bankers and traders from innovating at the brink of theory.
Check the sidebar of "The Money Illusion" for an introduction or Google market monetarism.
Relevant judgments here: Fiscal policy can't do anything monetary policy can't, so you should generally prefer QE to increased government debt, unless a separate case is made for government investment qua investment; with a properly behaving central bank the fiscal multiplier is exactly zero because the central bank is already stabilizing NGDP. Money is 'tight' or 'loose' in the general economy depending on NGDP growth rates; looking at interest rates or money supply will be exceedingly misleading as to what the effective policy is right now, never mind what it should be going forward.
-- Eliezer Yudkowsky
He predicted deflation as part of the economic crisis, which hasn't happened. He's mentioned that in several posts, most significantly in this one:
He also frequently writes about the state of macro economics, the politics involved, the different camps in the debate and what can be done about it. For example:
So here’s what should have happened: economists propounding these other
approaches should have said, “Gosh, I seem to have been wrong. I need to
rethink my approach.”
Oh, and by the way, I have done that. As I’ve written before, I rethought my
views about liquidity traps and currency crises after the Asian crisis of the
late 1990s; I rethought my views about advanced country debt and deficits
after making a wrong prediction in 2003 (although in that case my mistake was
in not taking my own model seriously enough).
And that same search will turn up many, many posts where he talks about intellectual honesty and validity of different arguments, ranging from "this guy is lying" to "I disagree, but this is a reasonable point of view".
One could make that argument about the ptolemaic system and epicycles. While krugman admits his prediction was wrong, it is a subjective argument as to whether what Krugman did in the end was to admit his model is really wrong, or if he decided it needed mere "adjustment". I know where I stand on that judgment.
I could also be biased: When I was a (math major) undergrad holding study sessions for lower-level mathematicians at a school with a very prestigious econ department, I recall the econ students coming to me with problems that seem to be specifically concocted to satisfy a culture that valued 'make a model such that you can take the gnarliest derivative possible so that you can remind everyone that you actually could pass calculus'. Gnarly derivatives, of course, rarely correspond to honestly measurable quantities (up to error). There was also NO dimensional analysis, so often equations would have the same term in an exponent, as a sum, and as a divisor.
In the real sciences for 90% of useful things, we take advantage of taylor's theorem and call everything locally linear.
Imagine how difficult astronomy would be if stars adjusted their spectra according to their anticipation of spectral trends, which was informed by emotion and by their reading of physics papers. Now imagine astronomers beaming spectral fashion magazines at the sun and being held responsible for skin cancer rates and crop yields. Imagine interest groups representing growers of different crops at different latitudes, as well as sunscreen industry groups, all beaming their own spectral fashion magazines at the sun. If that were the case, I doubt physicists would have had enough confidence in their understanding of the solar spectrum to deduce the existence of a new element from it .
There was an article on HN just the other day about Voyager 1 entering some unknown region of the universe. The physicists involved had no idea what was going on and all of their predictions turned out to be incorrect. So why no whiny article titled "Physics is a Lost Field"?
Macroeconomics is hard. It's nearly impossible to perform experiments validating large scale macroeconomic theories. If there were a way to accurately test the impact of the government borrowing and spending money, then it would be carried out. But realistically there's no feasible way to do that.
What's more, the author doesn't propose any solutions. Anybody can say this or that is screwed up. But that doesn't help anything. I'm sure the econ world would love to hear his theories that unquestionably solve the problems once and for all, but of course he didn't offer any.
Also, his mention of Freakonomics has me doubting his understanding of economics. The sumo example he mentions is a pretty straightforward application of microeconomics. He then goes on to say, "Freakonomics is, however, silent on monetary or fiscal policy." Monetary and fiscal policy are macroeconomic concepts. Economics is about making decisions and choosing between alternatives - how will people spend their money, what will they choose in this situation, etc. There's more to econ than just growing the national economy.
I'd say that the main problem is that economists are using numbers, and they really want to use them, but they're using them blindly. They'd love to be treated as physicists, but, sorry, they aren't yet. Economy is a very green field with little actual prediction capabilities (at least on a macro scale), and the sooner we accept it, the better.
And the bottom line: never ever trust anyone trying to justify the application of some economic policy based on mid-term or long-term predictions, no matter how many numbers may they throw at you.
The field simply is not a science. It has no predictive value.
And blaming Merton & Scholes for the Long-Term Capital Management fiasco is another error:
The fact that they (actually Meriwether, but anyway) invested where they should have not (Russian bonds, look at that! like investing in Greek bonds right now) is just an indicator that we are human. Also, it is quite disputable that the intervention was necessary. The Fed acted short-term (which is one of the big mistakes of modern governments) and it might have gone better than predicted.
You can just take a look at the performance of Renaissance Technologies along the years to realize that you can make a lot of money if you really invest (human & technological capital) on it. It is hard, but it can be done. And using maths & CS, little more.
When I say science, I just mean the gathering of knowledge through systematic, empirical means. I think that's what a lot of people mean by it. The problem with economics is that the empirical tools we have at our disposal are not very powerful, the data is noisy, and the factor-space is huge.
So when the Fed decides to drop the interest rates or a Harvard economics professor writes a report on how X will cause Y, we consider these to be absolutes. Meaning, if for every time X happens, there is a reactionary Y that needs to happen to keep growth consistent. We just believe because they've created an economy theory to support it, that is must be true.
This intellectual disagreement over the impact of fiscal decisions has spilled into the public domain with the fight between Nobel Prize winner Paul Krugman and the Harvard Duo of Carmen Reinhart and Kenneth Rogoff. Here is the letter written by Reinhart and Rogoff labeling Krugman‘s comments on their work as “spectacularly uncivil behavior.” The spat is over the impact of debt on economic growth.
What the author must have known, but fails to mention, is that Reinhart-Rogoff was wrong due their selective data usage and spreadsheet errors.
Why hopeless? Here are my guesses:
Data and Reality. Historically the people studying
economies just had far too little data
on their subject. By analogy they were
trying to understand, repair, or re-engineer
a car but had never looked under a car,
never popped the hood, didn't know what
piston rings were, and still were in doubt
if the thing had front wheel or rear wheel
drive. In a medical analogy, they had no
cadavers to dissect, X-rays, MRIs, blood
tests, etc. In an astronomical analogy,
they had no telescopes.
So, they never did well with what is usually
one of the first steps forward for a science
-- the descriptive part where we just give
a good description of the subject.
Astronomy, biology, thermodynamics,
chemistry, electricity and magnetism,
and more all started with good descriptions
of their subjects.
Or, for cars, start tinkering, as Henry
Ford did, with a lot of time with dirty
hands, and only later use finite element
analysis to build models of stress and
strain in continuum mechanics. Or
naval architects had a lot of experience
before they moved to towing tanks and
the Navier-Stokes equations.
Bluntly I had to conclude that the
academic economists really just didn't
have even a first, good descriptive
understanding of a real economy.
During WWII for war production planning,
Leontief worked on 'input-output' models of the
US economy. Good for him. But I was
told that the US academic economics community
very much did not like his work because it
was not 'theoretical' enough as in,
say, 'political economy'. So, it looked
like academic economics wanted to stay with
pomp, pretense, prestige, ignorance,
One of my Ph.D. advisors wanted me to
take a course in economics so that
if I did any work on a committee on
a 'public sector' problem, then I could defend
myself from attacks by floods of gibberish
from academic economists. I've had
no desire to do any such 'public sector'
work and have not, but I signed up for
the suggested course.
I wanted to be nice to the professor
and not cause trouble. So, during his
lecture with a lot of hand waving and
free hand curves but no data and nothing
convincing, I just took notes and said
nothing. Then after class I asked him
what he was assuming about his curves
-- continuity, uniform continuity,
differentiability, continuous differentiability,
monotonicity, concavity, pseudo-concavity,
quasi-concavity (e.g., in case he was intending
to use constraint qualifications for the
Kuhn-Tucker conditions for optimality).
He was unhappy. Later in the day, I
got a message to see my advisor. I was
out of the economics course -- the professor
claimed that I might disrupt the class.
That was not my intention, but good
riddance! But that professor's reaction
seemed to be a special case of a major
'feature' of academic economics --
have a tightly knit 'club' that
wants only true believers
and pushes out any skeptics. Or
the first rule of Economics Club
is never talk about the rules of
Net, academic economists know next to
nothing important about any real economy.
Real economic policy needs much more
in data on real economies, insight
into reality, good judgment,
and real effectiveness with applied
math than is common in academic
A number such as "unemployment", for example, is loaded. Are parents that could have a job but choose to stay home to take care of children "unemployed"?
So, start by observing, getting a lot of data,
and being just descriptive; economics
didn't and still doesn't do nearly enough
here. Then look a little deeper
and start to make some sense of it;
economics has done next to nothing
here, necessarily so since they
didn't get a good grade in the
first grade with the descriptive work.
For something with a little promise,
e.g., Leontief's work , they don't like
that and regard it as not
acceptable in Economics Club.
So, with such work, in astronomy
get Hertzsprung–Russell diagrams.
Now, continue: Take what is known about
weights in the periodic table and observe
that if press together two deuterium
atoms to make one helium 4 atom, then
lose some mass and, thus, get some energy.
Keep this up across the periodic table
and see that iron has the least such
energy. Observe that can keep
pressing light atoms together to get
heavier ones, e.g., carbon, oxygen,
..., iron and get off energy. Do a lot of
model building of the reactions in the
centers of distant stars. Check with
the data, e.g., the Hertzsprung–Russell diagrams.
Eventually come up with a well tested, apparently
quite solid, model of stars. A lot work,
bright ideas, spectroscopy, expansion rate of
the universe and red shift, nuclear physics,
etc. Didn't say it was easy.
Much the same in other fields of science that
have been successful.
So, if economics wants to copy that
methodology, then they need to get
their hands on more data. Then, say,
they need to look at the flows
that, intuitively, appear to be
generating the data. I know; I know;
too soon want a lot of curves on
propensities, and will have a tough
time there. Not guaranteed to be
easy. But neither were the other
sciences that were successful.
My view of academic economics is that they
sit in small, dark rooms with the doors
closed, hope for a single stroke
success comparable with E = mc^2 in
physics, dream up models, essentially
of imaginary economies, hope, but not
get very far. So, they don't want to
do the first, observational, low level
grunt work of data collection and
basic description that astronomy,
chemistry, biology, etc. did.
There have been suggestions that the
Princeton econ department has two
halves with one half, then with
Bernanke, the more empirical. If
so, good for the Fed and the US
since we will be less likely to be
stuck with crack pot ideas from
A danger here: So, with such work
with imaginary economies, some
to be defunct economist makes a lot
of absurd assumptions and finds some
optimal solution -- maybe he
studied some linear programming
Kuhn-Tucker theory (Arrow), Lagrange
multiplier theory (Debreu),
or dynamic programming
(Samuelson). Then, seeing a real
economy and an opportunity for
fame, status, prestige, tenure,
lots of undergraduate co-eds for
secretaries, he takes the solution
from his imaginary economy and
says that that is what the real
economy should do. Bummer.
Good way to kill tens of thousands
of people from unemployment,
busted homes, street crime,
infant mortality, drug abuse,
suicide, etc. Incompetence
is a bad thing; incompetent economists
are really ugly things.
Your point about unemployment is an example:
Early on I looked at it and asked,
where the heck is the definition, that is,
the empirical or operational definition?
Next, one level deeper, what the heck are
we really counting? E.g., might we be
partitioning the data in some way that
would be better? So here what do we
want for better? Sure, we want something
that can help us with reductionism,
that is, clear causality. But the
actual unemployment data is so poorly
defined and so "aggregated" that our
hope for anything causal is just
smoking funny stuff, like the astronomers
calling stars just points of light and
not looking deeper with spectroscopy,
the periodic table,
nuclear physics, etc.
Finally I understood what the empirical
economists intended to do with data such
as unemployment: They just said,
it's a measurement; hopefully it is done
in a consistent way month by month;
we agree we don't have a good definition
and that the data is too aggregated and
has basic problems in measurement,
but we just take the data as it is;
with the data, we will do just empirical
statistical modeling. Then, recognized or
not, they have essentially given up
making unemployment data an input
to anything reductionist, causal,
or scientific. Suckage. Meanwhile
academic economists are driving
late model cars, and millions of
people are suffering from unemployment
likely for no very good reasons.
The biggest questions to me are growth and sustainability. In the neoclassical world (I studied and degreed in the study) growth factors are all engogenous to economics:
The Neoclassical Growth Model, where output is a function of current stocks of capital and labor: Y = A K^α L^(1-α ), and capital accumulation is simply a function of investment and captial depreciation: K = sY - δK
The AK theory, which reduces growth to aggregate capital: Y = AK.
The product-variety model (Romer, Dixit & Stiglitz): production as a function of summed intermediate production functions based on capital
The Schumpeterian model, output is based on a productivity function (technology, assumed to be increasing without limit, and capital.
Technology increases. Inputs are perfectly substitutable. Innovation will produce substitutes.
And that's straight out of the first chapter of a graduate level text in economic growth (Aghion & Howitt, The Economics of Growth, ISBN 978-0-262-01263-8).
Then there's the whole class of heterodox economics:
Among these, the fields of thermoeconomics / biophysical economics / ecological economics, in which a, if not the primary factor relating to economic growth is energy. Many of the contributions to this field come from outside the field of economics, particularly from ecology, biology, and physics, but there are also a number of classically trained economists: Kenneth Boulding, Robert Ayres, Charles A.S. Hall, Robert Costanza, Herman Daly, H.T. Odum, and others.
Further work is being done by a wide range of people, some with training in finance and business, but "unbrainwashed" (in their own words) by economic orthodoxy. Gail Tverberg of Our Finite world (http://www.ourfiniteworld.com/) is key among these. Joesph Tainter has an extensive study of collapse of complex societies with profound implications for the Western European, now global, civilization.
When I look at the predictions and statements of mainstream economists, I see huge amounts of wishful and/or fuzzy thinking, and a long string of disappointments. When I look at the track record of the energy-oriented heterodoxy, I see a much stronger record. I tend to consider my own degree "economic woo" -- more similar to astrology and alchemy than astronomy and physics.
Not that it's _all_ bunk: markets are useful (though highly imperfect, and rare) concepts. Changing the relationship of money to underlying real wealth (which I'm increasingly convinced should be measured in fungible energy units, localized (and yes, that reads "FUEL"), and even predictions of how people generally _do_ behave (though not how they _should_) can be useful. But in planning and plotting a long-term path through the future, I'm increasingly fearful that economics is the wrong tool for the job, and that collectively we're going to have an Alan Greenspan moment, in which we find we've been grossly mislead by our belief system.
Ben Bernanke's Federal Reserve policies are largely the prescription of Milton Friedman and the monetarists. Friedman argued that the Great Depression was caused by allegedly tight monetary policy in the 1930's that resulted in a massive series of runs on banks and crashes. He argued that the Great Depression could have been averted by loose monetray policy along the lines of quantitative easing.
For many years, until recently, conservative, business, and libertarian groups embraced Friedman's ideas probably in part because quoted out of context, they shifted the blame for the Great Depression from misconduct and bad decisions by private corporations and wealthy individuals to the federal government and civil servants who provide little funding to conservative, business, and libertarian lobbying groups. Friedman's arguments also argued that the government need only have provided cheap money to prevent/cure the Great Depression rather than activist government programs such as Social Security and the alphabet soup of public works programs such as the Works Progress Administration (WPA).
Keynesian economics strongly disagreed with the Friedman/monetarist theory. Keynesians argue that the United Staes did have loose monetary policy in the 1930's and it did not work. They argue that the United States and much of the world was in a liquidity trap, an unusuall situation in which interest rates reach or nearly reach zero but a negative interest rate is needed to produce a revival of consumption and demand. Pouring money into the economy through the Federal Reserve or other central bank won't work. Nor will there be much inflation, because the money just sits in bank accounts unused.
In Keynesian economics, absent some extreme positive economic shock like the invention of a new energy source, the government must borrow heavily and spend heavily to restart the economy, pulling it out of the liquidity trap, which it is argued is what World War II finally did in the 1940s.
Keynesian economists like Paul Krugman and Dean Baker argue that the United States has been in a liquidity trap since the crash in 2008. The liquidity trap theory makes a prediction that has so far been borne out, that inflation will remain low despite the huge infusion of money from quantitative easing and huge budget deficits. These other folks such as John Taylor, Peter Schiff, Ron Paul, and various other cricits of both Ben Bernanke and the Keynesians like Krugman have been consistently wrong about inflation for the last five years.